Petroleum filtering device and use method

Abstract

The invention relates to the technical field of petroleum filtering, in particular to a petroleum filtering device and a using method.The petroleum filtering device comprises a filter body, a data collecting assembly and a data processor, and the data collecting assembly obtains petroleum state data of a liquid inlet and a liquid outlet of the filter body within a target time period; the data processor obtains suspected abnormal state data in the petroleum state data and corresponding suspected moments, and based on the suspected abnormal state data of all the suspected moments, target suspected pollution degrees of all the suspected moments are obtained; based on the numerical value change trend and the time continuity of the target suspected pollution degree of all the suspected moments, the pseudo-abnormal degree of the target time period is obtained, then the real pollution degree of the cut-off moment of the target time period is obtained, the next backwashing moment is obtained according to the real pollution degree, the accuracy of the backwashing moment is improved, and the backwashing efficiency is improved. And finally, the accuracy of backwashing starting control is improved, and the efficiency of the petroleum filtering device is improved.

Description

Technical Field

[0001] The invention relates to the technical field of petroleum filtering, and in particular to a petroleum filtering device and a use method thereof. Background Art

[0002] The oil filter is used to filter impurities from oil. In order to ensure the efficiency of impurity filtration, the oil filter is usually equipped with a flushing and sewage discharge device, or the oil filter and the flushing and sewage discharge device are integrated to form an automatic backwashing oil filter, which is used to perform regular or irregular backwashing of the oil filter to flush out the filtered impurities.

[0003] like Figure 1 As shown, a common structure of an oil filter device includes: an inlet 1, an outlet 2, a filter 3, a slag outlet 4 and a flushing guide valve 5. The existing automatic backwashing strategy is: set a fixed pressure difference threshold, and when the actual pressure difference data of the oil at the inlet 1 and the outlet 2 reaches the set pressure difference threshold, the backwashing is automatically started. However, the method of using a fixed pressure difference threshold for backwashing start control does not take into account the actual state changes of the oil in the oil filter device during the filtration process. For example, changes in oil temperature will change the viscosity of the oil medium, and the viscosity will affect the flow of the oil through the filter 3, resulting in a large error between the oil pressure difference data and the actual pressure difference data.

[0004] Therefore, in the backwashing method of the existing petroleum filter device, the backwashing start time is not accurately obtained, resulting in low backwashing start control accuracy, thereby affecting the operating efficiency of the petroleum filter device. Summary of the invention

[0005] In view of this, in order to solve the technical problem of inaccurate acquisition of backwash start time in the existing backwash start control method of the petroleum filter device, the present invention provides a petroleum filter device and a method of use.

[0006] The technical solutions adopted are as follows: The embodiment of the present invention provides a petroleum filter device, comprising a filter body, wherein the filter body comprises a liquid inlet and a liquid outlet, and the petroleum filter device further comprises: A data acquisition component, used for acquiring the oil status data of the liquid inlet and the liquid outlet within a target time period; A data processor is used to obtain suspected abnormal state data and corresponding suspected moments in the petroleum state data; based on the suspected abnormal state data at each suspected moment, obtain the target suspected contamination degree at each suspected moment; according to the numerical change trend and time continuity of the target suspected contamination degree at all suspected moments, obtain the pseudo-abnormal degree of the target time period; according to the pseudo-abnormal degree, obtain the real contamination degree at the last suspected moment in the target time period; according to the real contamination degree, obtain the next backwashing moment.

[0007] In some possible implementations, the oil status data includes pressure, temperature and flow rate; Obtaining suspected abnormal state data in the oil state data and the corresponding suspected time, including: The oil state data of the liquid inlet at each sampling time within the target time period is obtained, and the data is mapped to a three-dimensional coordinate system and clustered to obtain discrete points outside the cluster as first initial discrete points; the sampling time corresponding to the first initial discrete point is the first initial suspected time; The oil state data of the liquid outlet at each sampling time within the target time period is obtained, and the data is mapped to a three-dimensional coordinate system and clustered to obtain discrete points outside the cluster as second initial discrete points; the sampling time corresponding to the second initial discrete point is the second initial suspected time; The overlapping initial suspected moment between the first initial suspected moment and the second initial suspected moment is obtained, the overlapping initial suspected moment is the suspected moment, and the oil state data at the overlapping initial suspected moment is the suspected abnormal state data.

[0008] In some possible implementations, based on the suspected abnormal state data at each suspected moment, obtaining the target suspected pollution degree at each suspected moment includes: Based on the abnormal conditions of the pressure and flow rate of the liquid inlet and the liquid outlet in the suspected abnormal state data at each suspected moment, obtaining the initial suspected contamination degree at each suspected moment; Based on the correlation between the temperature change of the liquid inlet and the liquid outlet and the change of the initial suspected contamination degree in the initial suspected abnormal state data at each suspected moment, the initial suspected contamination degree is adjusted to obtain the target suspected contamination degree.

[0009] In some possible implementations, based on the abnormal conditions of the pressure and flow rate of the liquid inlet and the liquid outlet in the suspected abnormal state data at each suspected moment, obtaining the initial suspected contamination degree at each suspected moment includes: Obtaining discrete distances between the pressure of the suspected abnormal state data of the liquid inlet at the first suspected moment and the pressure characteristic positions of each of the clusters, and obtaining a first pressure discrete distance according to each discrete distance of the pressure; obtaining discrete distances between the flow of the suspected abnormal state data of the liquid inlet at the first suspected moment and the flow characteristic positions of each of the clusters, and obtaining a first flow discrete distance according to each discrete distance of the flow; the first suspected moment is any suspected moment; Obtaining the discrete distances between the pressure of the suspected abnormal state data of the liquid outlet at the first suspected moment and the pressure characteristic positions of each of the clusters, and obtaining the second pressure discrete distance according to each discrete distance of the pressure; obtaining the discrete distances between the flow rate of the suspected abnormal state data of the liquid outlet at the first suspected moment and the flow characteristic positions of each of the clusters, and obtaining the second flow discrete distance according to each discrete distance of the flow rate; Obtain a pressure discrete distance difference at a first suspected moment according to the first pressure discrete distance and the second pressure discrete distance, and obtain a flow discrete distance difference at a first suspected moment according to the first flow discrete distance and the second flow discrete distance; An initial suspected contamination degree at a first suspected moment is obtained according to the pressure discrete distance difference and the flow discrete distance difference.

[0010] In some possible implementations, the process of acquiring the correlation relationship includes: Obtaining the temperature difference between the first suspected moment and a preset number of suspected moments before the first suspected moment to form a temperature difference sequence; the temperature difference is the temperature difference in the suspected abnormal state data of the liquid inlet and the liquid outlet corresponding to the suspected moment; The correlation between the temperature difference sequence and the suspected pollution degree sequence at the first suspected moment is obtained; the suspected pollution degree sequence is composed of the suspected pollution degrees at the first suspected moment and the preset number of consecutive suspected moments before the first suspected moment.

[0011] In some possible implementations, the process of obtaining the suspected contamination level of the target includes: Acquire an adjustment coefficient according to the correlation degree at the first suspected moment, wherein the adjustment coefficient is negatively correlated with the correlation degree; The initial suspected pollution level is adjusted according to the adjustment coefficient to obtain a target suspected pollution level at the first suspected moment.

[0012] In some possible implementations, the numerical value change trend is specifically: the proportion of the number of suspected moments that meet the preset condition in the target suspected pollution degree sequence; the preset condition is: the target suspected pollution degree at the second suspected moment is greater than the target suspected pollution degree at the third suspected moment; the second suspected moment is any suspected moment, and the third suspected moment is the suspected moment before the second suspected moment; the target suspected pollution degree sequence is composed of the target suspected pollution degrees at all suspected moments; The process of acquiring the time continuity includes: acquiring the variance of the time interval sequence, wherein the time interval sequence is composed of the time intervals between all two adjacent sampling moments in the target suspected contamination degree sequence; and obtaining the time continuity according to the variance of the time interval sequence, wherein the time continuity is negatively correlated with the variance of the time interval sequence.

[0013] In some possible implementations, obtaining the true pollution degree at the last suspected moment in the target time period based on the pseudo-abnormality degree includes: According to the pseudo-abnormality degree, obtaining a perfection coefficient, wherein the perfection coefficient is negatively correlated with the pseudo-abnormality degree; The actual pollution degree is obtained according to the perfection coefficient and the target suspected pollution degree at the last suspected moment.

[0014] In some possible implementations, obtaining the next backwashing time according to the actual pollution degree includes: Obtaining a pollution degree difference between a preset real pollution degree threshold and the real pollution degree; The next backwashing time is obtained according to the pollution degree difference and the pollution degree change rate per unit time; the pollution degree change rate per unit time is calculated by the actual pollution degree and the characteristic time length of the target time period; the characteristic time length is the time interval between the start time of the target time period and the last suspected time.

[0015] A method for using a petroleum filtering device comprises the following steps: Obtaining the next backwashing time in the above-mentioned petroleum filtering device; When the next backwashing time comes, the flushing device equipped with the oil filter device is controlled to start so as to backwash the oil filter device.

[0016] The present invention has technical effects including but not limited to the following: the oil status data at the liquid inlet and the liquid outlet can reflect the actual operating status of the oil filtering device, based on the oil status data at the liquid inlet and the liquid outlet, and combined with the actual operating status of the oil filtering device in the target time period reflected in the oil status data, the next backwashing time closely related to the actual operating status is obtained, the accuracy of backwashing time acquisition is improved, and ultimately the accuracy of backwashing start-up control is improved, as well as the operating efficiency of the oil filtering device. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the mechanical structure of an existing petroleum filtering device; Figure 2 This is a schematic diagram of the electrical composition of a petroleum filtering device provided by the present invention; Figure 3 This is a specific electrical composition diagram of a petroleum filtering device provided by the present invention; Figure 4 It is a data processing flow chart of a petroleum filtering device provided by the present invention; Figure 5 It is a schematic diagram of a three-dimensional coordinate system; Figure 6 It is a schematic diagram of the process of obtaining suspected abnormal state data and the corresponding suspected time; Figure 7 It is a schematic diagram of the process of obtaining the suspected contamination degree of the target; Figure 8 It is a schematic diagram of the process of obtaining the initial suspected contamination level; Fig. 9 is a schematic diagram of the process of obtaining the correlation relationship; Fig.10 It is a schematic diagram of the process of obtaining the suspected contamination degree of the target; Fig.11 It is a schematic diagram of the process of obtaining the real pollution degree; Fig.12 It is a schematic diagram of the process of obtaining the next backwashing moment; Fig.13 It is an overall flow chart of the method of using a petroleum filtering device; In the figure, 1 is the liquid inlet, 2 is the liquid outlet, 3 is the filter screen, 4 is the slag discharge port, and 5 is the flushing guide valve. DETAILED DESCRIPTION

[0018] An embodiment of a petroleum filtering device: This embodiment provides a petroleum filtering device, including a filter body, the filter body such as Figure 1As shown, the filter body includes a liquid inlet 1 and a liquid outlet 2. Of course, the filter body also includes other components, such as: a filter screen 3, a slag discharge port 4 and a flushing guide valve 5. Usually, the petroleum filtering device is equipped with a flushing device. As a common component, the flushing device usually includes a flushing controller, a flushing motor and a flushing pump. The flushing controller is connected to the flushing motor and the flushing guide valve 5 by signal, and is used to send relevant control instructions to the flushing motor and the flushing guide valve 5. The flushing motor is connected to the flushing pump in a transmission manner, and the flushing pipeline of the flushing pump is connected to the flushing guide valve 5. The filtering process of the petroleum filtering device and the control of the flushing process are both existing technologies and will not be repeated here.

[0019] When the petroleum filtration device is in operation, the internal filtration efficiency will gradually deteriorate due to the accumulation of dirt and residue on the filter 3, resulting in a high pressure at the liquid inlet 1 and a low pressure at the liquid outlet 2. At this time, backwashing is required to remove the dirt on the filter 3. Normally, the accumulation of dirt has the characteristic of slowly increasing, that is, the data obtained is relatively continuous within a period of time after backwashing. However, when filtering petroleum, it is impossible to guarantee that the quality of the input oil is consistent during the monitoring stage, and there may be fluctuations, resulting in discontinuous data. Therefore, discontinuous data points may be due to high pollution.

[0020] like Figure 2 As shown, the petroleum filtering device also includes a data acquisition component and a data processor, and the data acquisition component and the data processor are signal connected.

[0021] The data acquisition component is used to obtain the oil state data of the liquid inlet 1 and the liquid outlet 2 within the target time period. Among them, the target time period is a preset time period, and the length of the time period is set according to actual needs. In this embodiment, in order to ensure the reliability of flushing, the starting time of the target time period is the starting time of the normal oil filtration after the last backwashing. Since the main purpose of the data processor is to judge the backwashing conditions of the oil filter device, remove pseudo-abnormal data, and accurately measure the next backwashing time of the oil filter device. Therefore, the end time of the target time period is earlier than the start time of the next backwashing. Therefore, the time length of the target time period is determined according to the time range of the time interval between two adjacent backwashings in the past, such as 25 minutes. Then, within 25 minutes after the end of the last backwashing, it is impossible to start the next backwashing. Even in extreme cases, after the end of the last backwashing, it is necessary to have an interval of at least more than 25 minutes before the next backwashing can be started.

[0022] The data acquisition component needs to ensure that the normal flow of oil inside the oil filter device is not affected, and the monitoring position of the data acquisition component in the liquid inlet 1 and the liquid outlet 2 is as close to the core flow area of ​​the oil as possible, so that the oil status data can be accurately monitored. Then, first of all, the data acquisition component can use a miniaturized detection device, and the data acquisition component can use wireless communication to transmit data to the data processor. For example: the data acquisition component is equipped with a low-power wireless communication module such as a Bluetooth module, and the data processor is also equipped with the same type of wireless communication module, so that the signal transmission line is not required, further avoiding affecting the normal operation of the oil filter device and reducing power consumption. Accordingly, a special power supply needs to be set in the data acquisition component, and a micro power supply such as a button battery can be used to reduce the volume and further avoid affecting the normal operation of the oil filter device.

[0023] As a specific implementation, the oil state data acquired by the data acquisition component are pressure, temperature and flow, wherein the pressure is the oil pressure, the temperature is the oil temperature, and the flow is the oil flow. Then, the data acquisition component includes a first data acquisition unit and a second data acquisition unit, such as Figure 3 As shown, the first data acquisition unit includes a first pressure sensor, a first temperature sensor and a first flow sensor, and the first data acquisition unit is arranged at the liquid inlet 1, and is used to detect the pressure, temperature and flow at the liquid inlet 1. The second data acquisition unit includes a second pressure sensor, a second temperature sensor and a second flow sensor, and the second data acquisition unit is arranged at the liquid outlet 2, and is used to detect the pressure, temperature and flow at the liquid outlet 2. It should be understood that each of the above sensors can be a small sensor, and the specific model is set according to actual needs.

[0024] It should be noted that the acquisition frequencies of the above sensors are consistent. For a certain sampling moment, the acquired data include the pressure, temperature and flow rate at the liquid inlet 1 and the pressure, temperature and flow rate at the liquid outlet 2 at the sampling moment.

[0025] The data processor is used to process data according to the oil status data to obtain the next backwashing time. The hardware device of the data processor can be configured as a data processing chip such as a CPU and a single-chip microcomputer.

[0026] The data processing strategy implemented by the data processor is as follows: Figure 4 As shown, the following steps are included: Step S1: Obtain suspected abnormal state data in the oil state data and the corresponding suspected time.

[0027] The oil state data can reflect the actual state of the oil during the filtering process. According to the actual state, the suspected abnormal state data can be obtained, thereby obtaining the suspected time corresponding to the suspected abnormal state data.

[0028] In this embodiment, a three-dimensional coordinate system is constructed, such as Figure 5 As shown, the X-axis of the three-dimensional coordinate system is temperature, the Y-axis is pressure, and the Z-axis is flow.

[0029] The process of obtaining suspected abnormal state data and the corresponding suspected time is as follows: Figure 6 As shown, including: Step S1-1: obtaining the oil state data of the liquid inlet at each sampling time within the target time period, mapping it to a three-dimensional coordinate system, and performing clustering to obtain discrete points outside the cluster as first initial discrete points; the sampling time corresponding to the first initial discrete point is the first initial suspected time; Step S1-2: obtaining the oil state data of the liquid outlet at each sampling time within the target time period, mapping it to a three-dimensional coordinate system, and clustering it to obtain discrete points outside the cluster as second initial discrete points; the sampling time corresponding to the second initial discrete point is the second initial suspected time; Step S1-3: obtaining the overlapping initial suspected moment between the first initial suspected moment and the second initial suspected moment, wherein the overlapping initial suspected moment is the suspected moment, and the oil state data at the overlapping initial suspected moment is the suspected abnormal state data.

[0030] For the oil status data of inlet 1: Since the target time period includes multiple sampling moments, the pressure, temperature and flow rate data at the liquid inlet 1 can be obtained at each sampling moment. Each sampling moment corresponds to a three-dimensional coordinate point, which is composed of the pressure, temperature and flow rate data at the corresponding sampling moment. Therefore, the three-dimensional coordinate points of the liquid inlet 1 at each sampling moment in the target time period are obtained, which is expressed as [ ],in, is the pressure at the xth sampling moment, is the temperature at the xth sampling moment, is the flow rate at the xth sampling moment. Then it is mapped into a three-dimensional coordinate system, so that there are multiple three-dimensional coordinate points in the three-dimensional coordinate system.

[0031] The three-dimensional coordinate points in the three-dimensional coordinate system are clustered. In this embodiment, the clustering algorithm is an existing algorithm, which is set according to actual needs, such as the K-means clustering algorithm, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm, etc. If the K-means clustering algorithm is used, the number of clusters, that is, the value of K, such as K=3, needs to be set in advance. The DBSCAN clustering algorithm does not need to set the number of clusters. This embodiment takes the DBSCAN clustering algorithm as an example.

[0032] The neighborhood size of DBSCAN clustering is set according to actual needs. As a specific implementation method, it is necessary to select a suitable neighborhood to ensure clustering performance during DBSCAN clustering. A neighborhood distance that is too large is likely to miscalculate discrete points into the normal interval, and a neighborhood that is too small is likely to treat normal points as discrete points. As data grows over time, the selected neighborhood size is the mean of adjacent data in space, so the selected neighborhood calculation method is: ; In the formula, represents the neighborhood selection size in DBSCAN, Indicates the average change in temperature during the target time period; Indicates the average change in flow rate during the target time period; Indicates the average change in pressure during the target time period. The calculation formula is generally expressed as the distance of the average change in three-dimensional space.

[0033] Among them, for any average change in temperature, flow rate and pressure, the calculation method is as follows: ; In the formula, represents the average change of the y-th type of oil state data, n represents the number of sampling moments in the target time period, represents the yth type of oil state data at the xth sampling moment, It represents the yth oil state data at the x+1th sampling moment; when y=1, 2 and 3, the yth oil state data is represented as temperature, pressure and flow rate respectively.

[0034] The number of clusters generated after clustering is determined by the actual clustering results. In the oil filtration process, the difference between the data in normal state is small, and it is easy to classify the normal state data into one category during clustering, while the data in suspected abnormal state is large in difference and easy to become isolated discrete points. Figure 5As shown, the coordinate points in the dotted ellipse area represent discrete points. Therefore, by clustering the oil state data of the liquid inlet 1, clusters and discrete points outside the clusters can be obtained, and the obtained discrete points are used as the first initial discrete points, and the sampling time corresponding to the first initial discrete points is the first initial suspected time.

[0035] The above process is adopted to perform the same clustering process on the petroleum state data of the liquid outlet 2, and the discrete points outside the cluster are obtained as the second initial discrete points. The sampling time corresponding to the second initial discrete points is the second initial suspected time.

[0036] Therefore, based on the liquid inlet 1, a plurality of first initial discrete points and their corresponding first initial suspected moments are obtained, and based on the liquid outlet 2, a plurality of second initial discrete points and their corresponding second initial suspected moments are obtained.

[0037] Comprehensively analyze the liquid inlet 1 and the liquid outlet 2 to obtain the overlapping initial suspected moment in the first initial suspected moment and the second initial suspected moment, that is, the same initial suspected moment in the first initial suspected moment and the second initial suspected moment. The overlapping initial suspected moment is used as the desired suspected moment, and the oil state data at the overlapping initial suspected moment is used as the desired suspected abnormal state data.

[0038] The suspected time and the corresponding suspected abnormal state data obtained in this step represent points that suddenly fluctuate compared to the overall situation, which do not conform to the characteristics of continuous and gradual increase in the pollution of the filter 3 and are suspected points with pollution characteristics.

[0039] Step S2: Based on the suspected abnormal state data at each suspected moment, obtain the target suspected contamination degree at each suspected moment.

[0040] The suspected abnormal state data at each suspected moment reflects the suspected contamination degree of the filter 3 at the corresponding suspected moment. Therefore, based on the suspected abnormal state data at each suspected moment, the target suspected contamination degree at each suspected moment is obtained. As a specific implementation method, Figure 7 As shown, a specific implementation process is given: Step S2-1: Based on the abnormal conditions of the pressure and flow rate of the liquid inlet and the liquid outlet in the suspected abnormal state data at each suspected moment, the initial suspected contamination degree at each suspected moment is obtained.

[0041] The severity of contamination of the filter 3 is usually reflected in the abnormal conditions of the pressure at the liquid inlet 1 and the liquid outlet 2, as well as the abnormal conditions of the flow rate. For example, the greater the pressure difference and flow rate difference between the liquid inlet 1 and the liquid outlet 2 at the same sampling moment, the higher the severity of contamination of the filter 3. Therefore, based on the abnormal conditions of the pressure and flow rate at the liquid inlet 1 and the liquid outlet 2 in the suspected abnormal state data at each suspected moment, the initial suspected contamination degree at each suspected moment is obtained.

[0042] Specifically, Figure 8 As shown in the figure, the process of obtaining the initial suspected contamination level includes: Step S2-1-1: obtaining the discrete distances between the pressure of the suspected abnormal state data of the liquid inlet at the first suspected moment and the pressure characteristic positions of each of the clusters, and obtaining the first pressure discrete distance according to each discrete distance of the pressure; obtaining the discrete distances between the flow rate of the suspected abnormal state data of the liquid inlet at the first suspected moment and the flow characteristic positions of each of the clusters, and obtaining the first flow discrete distance according to each discrete distance of the flow rate; the first suspected moment is any suspected moment; Step S2-1-2: obtaining the discrete distances between the pressure of the suspected abnormal state data of the liquid outlet at the first suspected moment and the pressure characteristic positions of each of the clusters, and obtaining a second pressure discrete distance according to each discrete distance of the pressure; obtaining the discrete distances between the flow rate of the suspected abnormal state data of the liquid outlet at the first suspected moment and the flow characteristic positions of each of the clusters, and obtaining a second flow discrete distance according to each discrete distance of the flow rate; Step S2-1-3: obtaining a pressure discrete distance difference at a first suspected moment according to the first pressure discrete distance and the second pressure discrete distance, and obtaining a flow discrete distance difference at a first suspected moment according to the first flow discrete distance and the second flow discrete distance; Step S2-1-4: Obtaining an initial suspected contamination degree at a first suspected moment according to the pressure discrete distance difference and the flow discrete distance difference.

[0043] For the sake of convenience, the first suspected moment is set to any suspected moment. For inlet 1: There may be only one or more clusters corresponding to the liquid inlet 1. Obtain the discrete distance between the pressure of the suspected abnormal state data at the first suspected moment of the liquid inlet 1 and the pressure characteristic position of each cluster corresponding to the liquid inlet 1. Among them, the pressure characteristic position of the cluster can be the pressure center position of the corresponding cluster. The calculation method of the pressure center position of the cluster is: obtain the average value of the pressure of all suspected abnormal state data in the cluster, and the pressure average value is the pressure center position of the cluster. Calculate the absolute value of the difference between the pressure of the suspected abnormal state data at the first suspected moment of the liquid inlet 1 and the pressure average value of each cluster as the pressure discrete distance corresponding to each cluster. Thus, the pressure discrete distance corresponding to each cluster is obtained, and then the average value of the pressure discrete distance of all clusters is calculated as the first pressure discrete distance at the first suspected moment of the liquid inlet 1.

[0044] Similarly, obtain the discrete distance between the flow rate of the suspected abnormal state data at the first suspected moment of the liquid inlet 1 and the flow characteristic position of each cluster corresponding to the liquid inlet 1. Among them, the flow characteristic position of the cluster can be the flow center position of the cluster. The calculation method of the flow center position of the cluster is: obtain the average value of the flow of all suspected abnormal state data in the cluster, and the flow average value is the flow center position of the cluster. Calculate the absolute value of the difference between the flow rate of the suspected abnormal state data at the first suspected moment of the liquid inlet 1 and the flow average value of each cluster corresponding to the liquid inlet 1 as the flow discrete distance corresponding to each cluster. In this way, the flow discrete distance corresponding to each cluster is obtained, and then the average value of the flow discrete distance of all clusters is calculated as the first flow discrete distance at the first suspected moment of the liquid inlet 1.

[0045] Similarly, using the above calculation process of the first pressure discrete distance and the first flow discrete distance, for the liquid outlet 2: the second pressure discrete distance and the second flow discrete distance at the first suspected moment of the liquid outlet 2 are obtained.

[0046] Then, the absolute value of the difference between the first pressure discrete distance and the second pressure discrete distance is calculated as the pressure discrete distance difference at the first suspected moment, and the absolute value of the difference between the first flow discrete distance and the second flow discrete distance is calculated as the flow discrete distance difference at the first suspected moment. The larger the pressure discrete distance difference, the greater the pressure difference between the liquid inlet 1 and the liquid outlet 2 at the first suspected moment, and the higher the degree of pollution may be. The larger the flow discrete distance difference, the greater the flow difference between the liquid inlet 1 and the liquid outlet 2 at the first suspected moment, and the higher the degree of pollution may be. Therefore, the initial suspected pollution degree at the first suspected moment is obtained according to the pressure discrete distance difference and the flow discrete distance difference, and the initial suspected pollution degree is positively correlated with the pressure discrete distance difference and the flow discrete distance difference. In this embodiment, the product of the pressure discrete distance difference and the flow discrete distance difference at the first suspected moment and the normalized result is used as the initial suspected pollution degree at the first suspected moment.

[0047] The normalization in this embodiment can adopt the existing related normalization methods, such as: , where Z is the normalized value, z is the input value, It is an exponential function with the natural constant e as its base.

[0048] Thus, the initial suspected contamination degree at each suspected moment is obtained.

[0049] Step S2-2: Based on the correlation between the temperature change of the liquid inlet and the liquid outlet in the initial suspected abnormal state data at each suspected moment and the change of the initial suspected contamination degree, the initial suspected contamination degree is adjusted to obtain a target suspected contamination degree.

[0050] Temperature is a key factor that affects changes in pressure and flow. There is a certain correlation between temperature and oil viscosity. When the temperature changes, the viscosity of the oil will also change. Generally speaking, when the temperature rises, the viscosity of the oil will decrease, resulting in an increasing trend in pressure and flow, which will further cause the degree of contamination to be higher than the actual level. Therefore, based on the correlation between the temperature changes of inlet 1 and outlet 2 and the initial suspected contamination level changes in the initial suspected abnormal state data at each suspected moment, the initial suspected contamination level is adjusted to obtain the target suspected contamination level.

[0051] like Fig. 9 As shown in FIG. 1 , the process of obtaining the relevant relationship includes: Step S2-2-1: obtaining the temperature difference between the first suspected moment and a preset number of suspected moments before it, forming a temperature difference sequence; the temperature difference is the temperature difference in the suspected abnormal state data of the liquid inlet and the liquid outlet at the corresponding suspected moment; Step S2-2-2: Obtain the correlation between the temperature difference sequence and the suspected pollution degree sequence at the first suspected moment; the suspected pollution degree sequence is composed of the suspected pollution degrees at the first suspected moment and the preset number of consecutive suspected moments before it.

[0052] Among them, the continuous preset number is set according to actual needs, such as 5, and the preset number of suspected moments are multiple consecutive suspected moments before the first suspected moment, and the last suspected moment among the preset number of suspected moments is the previous suspected moment adjacent to the first suspected moment. The temperature difference is the difference in temperature in the suspected abnormal state data of the liquid inlet 1 and the liquid outlet 2 corresponding to the suspected moment, for example: the temperature difference at the first suspected moment is the difference between the temperature in the suspected abnormal state data of the liquid inlet 1 at the first suspected moment and the temperature in the suspected abnormal state data of the liquid outlet 2 at the first suspected moment, and the temperature difference in this embodiment is specifically the absolute value of the temperature difference; the fourth suspected moment is set as any suspected moment among the continuous preset number of suspected moments before the first suspected moment, and the temperature difference at the fourth suspected moment is the difference between the temperature in the suspected abnormal state data of the liquid inlet 1 at the fourth suspected moment and the temperature in the suspected abnormal state data of the liquid outlet 2 at the fourth suspected moment. Thus, the temperature difference of the continuous preset number of suspected moments before the first suspected moment is obtained.

[0053] The temperature differences between the first suspected moment and a preset number of suspected moments before the first suspected moment constitute a temperature difference sequence corresponding to the first suspected moment in time sequence.

[0054] Since each suspected moment corresponds to a suspected pollution degree, the suspected pollution degrees of the first suspected moment and a preset number of consecutive suspected moments before it are obtained, and according to the suspected pollution degrees of the first suspected moment and a preset number of consecutive suspected moments before it, a suspected pollution degree sequence corresponding to the first suspected moment is constructed in time sequence.

[0055] The correlation between the temperature difference sequence at the first suspected moment and the suspected pollution degree sequence can be a Pearson correlation coefficient or a cosine similarity, which can be selected by the implementer according to the actual situation. As an optimization method, the temperature difference sequence and the suspected pollution degree sequence can be curve fitted respectively, and then the correlation between the two fitted curves can be obtained.

[0056] Then, according to the correlation of the first suspected moment, the initial suspected pollution level at the first suspected moment is adjusted to obtain the target suspected pollution level at the first suspected moment. Fig.10 As shown in the figure, the process of obtaining the suspected contamination level of the target includes: Step S2-2-3: obtaining an adjustment coefficient according to the correlation degree at the first suspected moment, wherein the adjustment coefficient is negatively correlated with the correlation degree; Step S2-2-4: Adjust the initial suspected pollution level according to the adjustment coefficient to obtain the target suspected pollution level at the first suspected moment.

[0057] As the temperature increases, the pressure and flow rate will tend to increase, further causing the pollution level to be higher than the actual level. Therefore, if the temperature change is more consistent with the difference change in the suspected pollution level, that is, the higher the correlation between the temperature difference sequence and the suspected pollution level sequence at the first suspected moment, then it is considered that the possibility of data anomaly caused by temperature is higher, and the actual pollution level is lowered accordingly. Therefore, the adjustment coefficient is obtained according to the correlation level at the first suspected moment, and the adjustment coefficient is negatively correlated with the correlation level. It should be understood that the adjustment coefficient is a value less than or equal to 1, and the adjustment coefficient must be within a certain range, otherwise the target suspected pollution level obtained after adjustment may be too small, thereby affecting the accuracy. Therefore, as a specific implementation method, a specific quantification method of the adjustment coefficient is given as follows: ; in, represents the adjustment coefficient of the i-th suspected moment, represents the correlation degree of the i-th suspected moment, represents the normalized function. The i-th suspected moment represents any suspected moment.

[0058] Then, the adjustment coefficient at the first suspected moment is multiplied by the initial suspected pollution level at the first suspected moment, and the obtained product is the target suspected pollution level at the first suspected moment.

[0059] Thus, the target suspected pollution degree at each suspected moment is obtained.

[0060] Step S3: According to the numerical variation trend and time continuity of the target suspected pollution degree at all suspected moments, the pseudo abnormality degree of the target time period is obtained.

[0061] The target suspected pollution levels at all suspected moments are arranged in time sequence to form a target suspected pollution level sequence. The actual abnormality of the target time period is related to the numerical change trend and time continuity of the target suspected pollution level sequence. Therefore, the pseudo abnormality of the target time period can be obtained based on the numerical change trend and time continuity of the target suspected pollution level sequence.

[0062] As a specific implementation, the numerical change trend characterizes the numerical rising trend of the target suspected pollution degree. The numerical change trend is specifically: the proportion of the number of suspected moments that meet the preset conditions in the target suspected pollution degree sequence. The preset condition is: the target suspected pollution degree at the second suspected moment is greater than the target suspected pollution degree at the third suspected moment; the second suspected moment is any suspected moment, and the third suspected moment is the previous suspected moment of the second suspected moment. The target suspected pollution degree at the second suspected moment is greater than the target suspected pollution degree at the third suspected moment, which means that it is greater than the target suspected pollution degree at the adjacent previous suspected moment. Then, the numerical change trend is essentially the proportion of the number of suspected moments that are greater than the target suspected pollution degree at the adjacent previous suspected moment, that is, the ratio of the obtained number to the total number of suspected moments. As another implementation, the following calculation can also be performed: obtain the difference between the previous target suspected pollution degree and the target suspected pollution degree at the next similar moment in the two adjacent suspected moments in the target suspected pollution degree sequence, and then obtain the number of differences greater than 0 from the differences between all two adjacent suspected moments. The difference greater than 0 indicates that the previous target suspected pollution degree is greater than the target suspected pollution degree at the next similar moment.

[0063] The process of obtaining time continuity is as follows: obtain the sampling moments corresponding to each target suspected pollution degree in the target suspected pollution degree sequence to form a sampling moment sequence; obtain the time interval between each two adjacent sampling moments in the sampling moment sequence, and the time intervals between all two adjacent sampling moments constitute a time interval sequence; then, obtain the time interval variance of the time interval sequence. The smaller the time interval variance, the higher the time continuity of the target suspected pollution degree sequence. Therefore, the time interval variance is negatively correlated and normalized in the end, and the result is time continuity.

[0064] The negative correlation normalization in this embodiment can adopt the existing correlation normalization method, such as: , where Z is the normalized value of negative correlation, z is the input value, It is an exponential function with the natural constant e as its base.

[0065] The normal contamination of the filter 3 is that the dirt accumulation is slow and rising. The more normal the contamination is, the higher the time continuity is and the higher the numerical change trend is. Therefore, according to the numerical change trend and time continuity of the target suspected contamination degree at all suspected moments, the pseudo abnormal degree of the target time period is obtained. As a specific implementation method, the product of the numerical change trend and the time continuity is used as the pseudo abnormal degree of the target time period.

[0066] Step S4: According to the pseudo-abnormality degree, the actual pollution degree at the last suspected moment in the target time period is obtained.

[0067] According to the obtained pseudo-abnormal degree of the target time period, the real pollution degree of the last suspected moment in the target time period can be obtained. The real pollution degree is finally used to obtain the next backwashing moment.

[0068] As a specific implementation, Fig.11 As shown, a specific calculation process of the actual pollution degree is given as follows: Step S4-1: obtaining a perfection coefficient according to the pseudo-abnormality degree, wherein the perfection coefficient is negatively correlated with the pseudo-abnormality degree; Step S4-2: Obtain the actual pollution level according to the improvement coefficient and the target suspected pollution level at the last suspected moment.

[0069] Specifically, the last suspected moment in the target time period is obtained from the target time period, and the target suspected pollution degree at the last suspected moment is obtained accordingly. The improvement coefficient is used as an adjustment coefficient of the target suspected pollution degree at the last suspected moment, and is obtained from the pseudo-abnormality degree. The improvement coefficient is negatively correlated with the pseudo-abnormality degree. As a specific implementation method, the difference between the value 1 and the pseudo-abnormality degree is used as the improvement coefficient. Finally, according to the improvement coefficient and the target suspected pollution degree at the last suspected moment, the real pollution degree at the last suspected moment is obtained. As a specific implementation method, the product of the improvement coefficient and the target suspected pollution degree at the last suspected moment is used as the real pollution degree at the last suspected moment. As time goes by, the pollution of the filter 3 becomes more serious and the pollution situation becomes more complicated. Therefore, the last suspected moment of the target time period is very close to or even equal to the end time of the target time period. Therefore, the real pollution degree at the last suspected moment is characterized as the real pollution degree of the filter 3 at the end of the target time period.

[0070] Step S5: Obtain the next backwashing time according to the actual pollution degree.

[0071] According to the actual pollution degree at the last suspected moment, the next backwashing moment is obtained. As a specific implementation method, Fig.12 As shown, a specific process for obtaining the next backwashing time is given as follows: Step S5-1: Obtaining a pollution degree difference between a preset real pollution degree threshold and the real pollution degree; Step S5-2: Obtain the next backwashing time according to the pollution degree difference and the pollution degree change rate per unit time; the pollution degree change rate per unit time is calculated by the actual pollution degree and the characteristic time length of the target time period; the characteristic time length is the time interval between the start time of the target time period and the last suspected time.

[0072] A real contamination threshold is preset, and the preset real contamination threshold is obtained by experience. As a specific example, it is set to 0.64. The preset real contamination threshold represents the critical value of contamination that can be used for the next backwashing.

[0073] The difference in pollution degree between the preset real pollution degree threshold and the real pollution degree indicates the difference between the real pollution degree and the preset real pollution degree threshold. In this embodiment, the difference between the preset real pollution degree threshold and the real pollution degree is used as the pollution degree difference. The pollution degree unit time change rate indicates the change of the pollution degree in unit time. In this embodiment, the pollution degree unit time change rate is calculated by the real pollution degree and the characteristic time length of the target time period. As a specific implementation method, the ratio of the real pollution degree to the characteristic time length of the target time period is used as the pollution degree unit time change rate. The characteristic time length is the time interval between the starting moment of the target time period and the last suspected moment. It should be noted that if the actual scene is not considered, in principle, there is a possibility that the starting moment of the target time period and the last suspected moment are the same moment, that is, the obtained time interval is 0. However, in the actual oil filtration process, the later the time is, the pollutants gradually accumulate on the filter 3, the more serious the pollution of the filter 3, and the more complicated the pollution situation is, that is, the later the time is, the more suspected moments may be, then there will inevitably be multiple suspected moments in the target time period, and there is no situation where the above time interval is 0. In addition, even if the time interval is 0, the process of obtaining the next backwashing time in this embodiment is not implemented in this case.

[0074] Finally, according to the difference in pollution degree and the rate of change of pollution degree per unit time, the time interval between the last suspected moment and the next backwashing moment of the target time period is obtained. Specifically, the product of the difference in pollution degree and the rate of change of pollution degree per unit time is used as the time interval between the last suspected moment and the next backwashing moment. The time interval indicates how long it takes to reach the preset real pollution degree threshold after the last suspected moment, that is, how long it takes to start backwashing. Thus, the next backwashing moment is obtained according to the time interval. It should be understood that if the time interval between the last suspected moment and the next backwashing moment is less than or equal to the time interval between the last suspected moment in the target time period and the end time of the target time period, that is, the next backwashing moment obtained is earlier than or equal to the end time of the target time period, then the end time of the target time period is used as the next backwashing moment.

[0075] Subsequently, after obtaining the next backwashing time, the data processor uses a timer to output relevant control instructions when the next backwashing time arrives, so as to perform backwashing control.

[0076] An embodiment of a method for using a petroleum filtering device: This embodiment provides a method for using a petroleum filter device. The object of the method is the petroleum filter device in the above-mentioned petroleum filter device embodiment. The specific data required is the next backwashing time obtained in the above-mentioned petroleum filter device embodiment. Fig.13 As shown, the process of using this method is: Step S100: obtaining the next backwashing time in the petroleum filtering device; Step S200: When the next backwashing time arrives, the flushing device equipped with the oil filter device is controlled to start up to backwash the oil filter device.

[0077] The oil filter device is equipped with a flushing device, and the data processor is connected to the flushing device by signal. When the next backwashing time arrives, the data processor sends a start instruction to the flushing device, specifically: the data processor outputs a start instruction to the flushing controller in the flushing device, and the flushing controller outputs a control instruction to the flushing motor and the flushing guide valve 5 to control the start of the flushing motor and the conduction of the flushing guide valve 5 to backwash the oil filter device. It should be understood that when the next backwashing time arrives or is about to arrive, the data processor also sends a shutdown instruction to the oil filter device to stop oil filtration to ensure reliable backwashing.

[0078] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A petroleum filtering device, comprising a filter body, wherein the filter body comprises a liquid inlet and a liquid outlet, wherein: The petroleum filtering device also includes: A data acquisition component, used for acquiring the oil status data of the liquid inlet and the liquid outlet within a target time period; A data processor is used to obtain suspected abnormal state data and corresponding suspected moments in the petroleum state data; based on the suspected abnormal state data at each suspected moment, obtain the target suspected contamination degree at each suspected moment; according to the numerical change trend and time continuity of the target suspected contamination degree at all suspected moments, obtain the pseudo-abnormal degree of the target time period; according to the pseudo-abnormal degree, obtain the real contamination degree at the last suspected moment in the target time period; according to the real contamination degree, obtain the next backwashing moment.

2. The petroleum filtering device according to claim 1, characterized in that: The oil status data includes pressure, temperature and flow rate; Obtaining suspected abnormal state data in the oil state data and the corresponding suspected time, including: The oil state data of the liquid inlet at each sampling time within the target time period is obtained, and the data is mapped to a three-dimensional coordinate system and clustered to obtain discrete points outside the cluster as first initial discrete points; the sampling time corresponding to the first initial discrete point is the first initial suspected time; The oil state data of the liquid outlet at each sampling time within the target time period is obtained, and the data is mapped to a three-dimensional coordinate system and clustered to obtain discrete points outside the cluster as second initial discrete points; the sampling time corresponding to the second initial discrete point is the second initial suspected time; The overlapping initial suspected moment between the first initial suspected moment and the second initial suspected moment is obtained, the overlapping initial suspected moment is the suspected moment, and the oil state data at the overlapping initial suspected moment is the suspected abnormal state data.

3. The petroleum filtering device according to claim 2, characterized in that: Based on the suspected abnormal state data at each suspected moment, the target suspected pollution degree at each suspected moment is obtained, including: Based on the abnormal conditions of the pressure and flow rate of the liquid inlet and the liquid outlet in the suspected abnormal state data at each suspected moment, obtaining the initial suspected contamination degree at each suspected moment; Based on the correlation between the temperature change of the liquid inlet and the liquid outlet and the change of the initial suspected contamination degree in the initial suspected abnormal state data at each suspected moment, the initial suspected contamination degree is adjusted to obtain the target suspected contamination degree.

4. The petroleum filtering device according to claim 3, characterized in that: Based on the abnormal conditions of the pressure and flow rate of the liquid inlet and the liquid outlet described in the suspected abnormal state data at each suspected moment, the initial suspected contamination degree at each suspected moment is obtained, including: Obtaining discrete distances between the pressure of the suspected abnormal state data of the liquid inlet at the first suspected moment and the pressure characteristic positions of each of the clusters, and obtaining a first pressure discrete distance according to each discrete distance of the pressure; obtaining discrete distances between the flow of the suspected abnormal state data of the liquid inlet at the first suspected moment and the flow characteristic positions of each of the clusters, and obtaining a first flow discrete distance according to each discrete distance of the flow; the first suspected moment is any suspected moment; Obtaining the discrete distances between the pressure of the suspected abnormal state data of the liquid outlet at the first suspected moment and the pressure characteristic positions of each of the clusters, and obtaining the second pressure discrete distance according to each discrete distance of the pressure; obtaining the discrete distances between the flow rate of the suspected abnormal state data of the liquid outlet at the first suspected moment and the flow characteristic positions of each of the clusters, and obtaining the second flow discrete distance according to each discrete distance of the flow rate; Obtain a pressure discrete distance difference at a first suspected moment according to the first pressure discrete distance and the second pressure discrete distance, and obtain a flow discrete distance difference at a first suspected moment according to the first flow discrete distance and the second flow discrete distance; An initial suspected contamination degree at a first suspected moment is obtained according to the pressure discrete distance difference and the flow discrete distance difference.

5. The petroleum filtering device according to claim 4, characterized in that: The process of obtaining the correlation relationship includes: Obtaining the temperature difference between the first suspected moment and a preset number of suspected moments before the first suspected moment to form a temperature difference sequence; the temperature difference is the temperature difference in the suspected abnormal state data of the liquid inlet and the liquid outlet corresponding to the suspected moment; The correlation between the temperature difference sequence and the suspected pollution degree sequence at the first suspected moment is obtained; the suspected pollution degree sequence is composed of the suspected pollution degrees at the first suspected moment and the preset number of consecutive suspected moments before the first suspected moment.

6. The petroleum filtering device according to claim 5, characterized in that: The process of obtaining the suspected contamination degree of the target includes: Acquire an adjustment coefficient according to the correlation degree at the first suspected moment, wherein the adjustment coefficient is negatively correlated with the correlation degree; The initial suspected pollution level is adjusted according to the adjustment coefficient to obtain a target suspected pollution level at the first suspected moment.

7. The petroleum filtering device according to claim 1, characterized in that: The numerical variation trend is specifically: the proportion of the number of suspected moments that meet the preset conditions in the target suspected pollution degree sequence; the preset condition is: the target suspected pollution degree at the second suspected moment is greater than the target suspected pollution degree at the third suspected moment; the second suspected moment is any suspected moment, and the third suspected moment is the suspected moment before the second suspected moment; the target suspected pollution degree sequence is composed of the target suspected pollution degrees at all suspected moments; The process of acquiring the time continuity includes: acquiring the variance of the time interval sequence, wherein the time interval sequence is composed of the time intervals between all two adjacent sampling moments in the target suspected contamination degree sequence; and obtaining the time continuity according to the variance of the time interval sequence, wherein the time continuity is negatively correlated with the variance of the time interval sequence.

8. The petroleum filtering device according to claim 1, characterized in that: Based on the pseudo-abnormality level, obtaining the real pollution level at the last suspected moment in the target time period includes: According to the pseudo-abnormality degree, obtaining a perfection coefficient, wherein the perfection coefficient is negatively correlated with the pseudo-abnormality degree; The actual pollution degree is obtained according to the perfection coefficient and the target suspected pollution degree at the last suspected moment.

9. The petroleum filtering device according to claim 1, characterized in that: According to the actual pollution degree, the next backwashing time is obtained, including: Obtaining a pollution degree difference between a preset real pollution degree threshold and the real pollution degree; The next backwashing time is obtained according to the pollution degree difference and the pollution degree change rate per unit time; the pollution degree change rate per unit time is calculated by the actual pollution degree and the characteristic time length of the target time period; the characteristic time length is the time interval between the start time of the target time period and the last suspected time.

10. A method for using a petroleum filtering device, characterized in that: The steps include: Obtaining the next backwashing time in the petroleum filtering device according to any one of claims 1 to 9; When the next backwashing time comes, the flushing device matched with the oil filter device is controlled to start so as to backwash the oil filter device.

Images