A method for adaptively adjusting the frequency of sensor cleaning brushes

Abstract

The present invention provides a method for adaptively adjusting the frequency of a sensor cleaning brush, including a data collection step, a data calculation and comparison step, and a cleaning cycle modification step. In the current cleaning cycle, a preset number of sensor detection data are collected; the stability value of the full amount of data in the current cleaning cycle is calculated and compared with the preset stability standard value, thereby obtaining a corrected cleaning frequency value for correcting the next cleaning cycle. By adopting the technical solution of the present invention, the frequency of the sensor cleaning brush is a non-fixed frequency, and the cleaning frequency is automatically adjusted according to environmental changes, which can effectively improve the cleaning effect of the cleaning brush and reduce the influence of external environmental factors on the measurement accuracy of the sensor; when the outer diameter environment is better, the cleaning frequency is automatically reduced, which can effectively extend the service life of the cleaning brush.

Description

Technical Field

[0001] The invention relates to the technical field of water quality detection sensors, and in particular to a method for adaptively adjusting the frequency of a sensor cleaning brush. Background Art

[0002] At present, water quality sensors are widely used and can be used in different water environments. In order to improve the detection accuracy and promptly remove the influence of impurities on the detection results, many water quality sensors come with cleaning brushes. For example, in the patent document with patent announcement number CN205757995U, the applicant proposed a cleaning brush and a device for automatically controlling the operation of the cleaning brush. However, the cleaning brushes of existing water quality sensors usually operate at a fixed frequency, and usually set different start-up frequencies for different environments. However, the actual environment is constantly changing over time, and the set fixed frequency cleaning cannot adapt to the changing environmental conditions. This may result in insufficient cleaning frequency and failure to achieve normal cleaning effects, thereby affecting the accuracy of the test data, or excessive cleaning frequency and reducing the product life. Summary of the invention

[0003] In order to overcome the above-mentioned deficiencies of the prior art, an object of the present invention is to provide a method for adaptively adjusting the frequency of a sensor cleaning brush.

[0004] In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention to solve the technical problem is: a method for adaptively adjusting the frequency of a sensor cleaning brush, comprising:

[0005] Data collection step: taking the period between two consecutive cleaning starts of the cleaning brush as a cleaning cycle, and collecting sensor detection data for a preset number of times within the cleaning cycle;

[0006] Data calculation and comparison steps: after the data collection is completed, the stability value of the full amount of data in this cleaning cycle is calculated and compared with the preset stability standard value; when the actual measured stability value of the full amount of data is worse than the preset stability standard value, the cleaning frequency is increased; when the actual measured stability value of the full amount of data is better than the preset stability standard value, the cleaning frequency is reduced to obtain a corrected cleaning frequency; when the actual measured stability value of the full amount of data is within the preset stability standard value interval, the cleaning frequency remains unchanged; thereby obtaining a corrected cleaning frequency value for correcting the next cleaning cycle;

[0007] Cleaning cycle modification steps: After the cleaning brush starts cleaning again, the duration of the new cleaning cycle is modified based on the corrected cleaning frequency value obtained in the previous cleaning cycle; in the new cleaning cycle, the sensor detection data is collected for the same preset number of times, and a new corrected cleaning frequency is calculated to modify the duration of the next cleaning cycle.

[0008] By adopting the technical solution of the present invention, the frequency of the sensor cleaning brush is a non-fixed frequency, and the cleaning frequency is automatically adjusted according to environmental changes, which can effectively improve the cleaning effect of the cleaning brush and reduce the impact of external environmental factors on the measurement accuracy of the sensor; when the outer diameter environment is better, the cleaning frequency is automatically reduced, which can effectively extend the service life of the cleaning brush.

[0009] Further, after the sensor is powered on, the cleaning brush starts cleaning for the first time, and the cleaning cycle duration between the first cleaning start and the second cleaning start is determined according to the preset cleaning frequency.

[0010] By adopting the above preferred solution, a cleaning frequency suitable for the applied water environment is preset in the sensor to ensure stable startup.

[0011] Furthermore, in the data calculation and comparison step, the stability value of the full amount of data is calculated according to the variance formula to obtain the variance S of the full amount of data; the preset stability standard value is the variance upper limit value S S1 , the lower limit of variance is S S2 ; When S S2 When S>S S1 When S S2 ≤S≤S S1 The cleaning frequency remains unchanged.

[0012] By adopting the above preferred scheme and comparing the variance, the stability state of the data can be reliably evaluated.

[0013] Furthermore, in the data collection step, within the current cleaning cycle of duration T, the preset number of sensor detection data collection is K, and the K data are divided into m groups according to the order of collection time, each group contains n data, where K = m*n;

[0014] In the data calculation and comparison step, when S>S S1 When the mean of K data is used to calculate the variance of each group of data in m groups, the variance of the first group of data is S1, the variance of the second group of data is S2, and so on until the variance of the mth group of data is S m , S 1、 S2…S m Sequentially with the upper limit of variance S S1 For comparison, if the variance S of the tth group of data is t Start larger than S S1 , then the next cleaning cycle is modified to (t-1)*T / m, where 1≤t≤m.

[0015] By adopting the above preferred solution, through data grouping, it is possible to further mine the source of unstable data and quickly adjust the cleaning cycle to the optimal state.​

[0016] Furthermore, in the data calculation and comparison step, when S S2 When the next cleaning cycle is changed to a*T, a is a constant, 1 <a≤2。

[0017] By adopting the above preferred solution, when the environment is better, the cleaning cycle is gradually expanded, which is conducive to the steady progress of adaptive adjustment.

[0018] Furthermore, the number of groups m of K data satisfies: [k 1 / 2 ]-1≤m≤[k 1 / 2 ]+1.

[0019] By adopting the above preferred solution, both the data volume and adjustment accuracy of each group are taken into consideration, thereby ensuring the reliability of the analysis data. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 creative work.

[0021] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] like Figure 1 As shown, in one embodiment of the present invention, a method for adaptively adjusting the frequency of a sensor cleaning brush includes:

[0024] Data collection step: taking the period between two consecutive cleaning starts of the cleaning brush as a cleaning cycle, and collecting sensor detection data for a preset number of times within the cleaning cycle;

[0025] ​Data calculation and comparison steps: after data collection is completed, calculate the stability value of the full amount of data in this cleaning cycle and compare it with the preset stability standard value; when the actual measured stability value of the full amount of data is worse than the preset stability standard value, increase the cleaning frequency; when the actual measured stability value of the full amount of data is better than the preset stability standard value, reduce the cleaning frequency to obtain the corrected cleaning frequency; when the actual measured stability value of the full amount of data is within the preset stability standard value range, the cleaning frequency remains unchanged;

[0026] Cleaning cycle modification steps: After the cleaning brush starts cleaning again, the duration of the new cleaning cycle is modified based on the corrected cleaning frequency obtained in the previous cleaning cycle; in the new cleaning cycle, the sensor detection data is collected for the same preset number of times, and a new corrected cleaning frequency is calculated to modify the duration of the next cleaning cycle.

[0027] The beneficial effects of adopting the above technical solution are: the frequency of the sensor cleaning brush is a non-fixed frequency, and the cleaning frequency is automatically adjusted according to environmental changes, which can effectively improve the cleaning effect of the cleaning brush and reduce the impact of external environmental factors on the measurement accuracy of the sensor; when the outer diameter environment is better, the cleaning frequency is automatically reduced, which can effectively extend the service life of the cleaning brush.

[0028] In other embodiments of the present invention, after the sensor is powered on, the cleaning brush starts cleaning for the first time, and the cleaning cycle duration between the first cleaning start and the second cleaning start is determined according to the preset cleaning frequency. The beneficial effect of the above technical solution is that the sensor presets a cleaning frequency that is suitable for the application water environment to ensure stable startup.

[0029] In some other embodiments of the present invention, in the data calculation and comparison step, the stability value of the full amount of data is calculated according to the variance formula to obtain the variance S of the full amount of data; the preset stability standard value is the variance upper limit value S S1 , the lower limit of variance is S S2 ; When S S2 When S>S S1 When S S2 ≤S≤S S1 The beneficial effect of adopting the above technical solution is that the stability state of the data can be reliably evaluated through variance calculation and comparison.

[0030] In some other embodiments of the present invention, a method for adaptively adjusting the frequency of a sensor cleaning brush includes:

[0031] Data collection steps: In the current cleaning cycle of duration T, that is, the current cleaning frequency is 1 / T, and the preset number of sensor detection data collection is K; ​

[0032] Data calculation and comparison steps: According to the chronological order of acquisition time, K data are divided into m groups, with each group containing n data, where K = m * n; the stability value of the full amount of data is calculated according to the variance formula to obtain the variance S of the full amount of data; the preset stability standard value is the variance upper limit value S S1 , and the variance lower limit value is S S2 ; when S > S S1 , the mean value of the K data is used to calculate the variance of each group of data in the m groups respectively, and the variance of the first group of data is obtained as S1, the variance of the second group of data is S2, until the variance of the mth group of data is S m , and S 1、 S2…S m are compared with the variance upper limit value S S1 in turn. If the variance S t of the tth group of data starts to be greater than S S1 , then the next cleaning cycle is modified to (t - 1) * T / m, where 1 ≤ t ≤ m; when S < S S2 , the next cleaning cycle is modified to a * T, where a is a constant and 1 < a ≤ 2; when S S2 ≤ S ≤ S S1 , the cleaning frequency remains unchanged; thus, the corrected cleaning frequency value for modifying the next cleaning cycle is obtained;

[0033] Cleaning cycle modification steps: After the cleaning brush starts cleaning again, based on the corrected cleaning frequency value obtained in the previous cleaning cycle, the duration of the new cleaning cycle is modified; within the new cleaning cycle, the sensor detection data is also collected a preset number of times, and the new corrected cleaning frequency value used to modify the duration of the next cleaning cycle is calculated.

[0034] In some other embodiments of the present invention, the number of groups m of the K data satisfies: [k 1 / 2 -1 ≤ m ≤ [k 1 / 2 +1. The beneficial effect of adopting the above technical solution is: taking into account the data volume of each group and the adjustment accuracy, and ensuring the reliability of the analysis data.

[0035] The following is an example to illustrate the adaptive adjustment process of the present invention:

[0036] 1. After the sensor is powered on, the cleaning brush starts cleaning for the first time. Assuming the initial cleaning cycle is once every 30 minutes, the single-chip microcomputer internally opens a buffer, and the preset buffer is 30 numbers, with a buffer filling frequency of 1 minute / number. Therefore, the time required to fill the buffer is 30 minutes. In the algorithm, the buffer is set as a two-dimensional matrix array, buffer[m][n], where m is 5 and n is 6, and the array is as follows:

[0037]

[0038] The sensor fills the buffer with data every minute. The startup data is buffer[0][0], the first minute is buffer[0][1], the fourth minute is buffer[0][5], and so on. The final buffer value is Buffer[4][5]. In the first 30 minutes after startup, there will be 5 groups of valid data with m=5, and each group contains 6 valid data with n=6.

[0039] 2. After the data filling is completed, analyze the existing data. First, calculate it according to the variance formula. The formula is as follows:

[0040]

[0041] According to the formula, we can get the variance S of the full amount of data, and the total amount of data is 30.

[0042] At the same time, the mean X value of the 30 data is used to calculate the variance S1 of buffer[0][0]~buffer[0][5]. Similarly, the variance S2 of buffer[1][0]~buffer[1][5] corresponding to m=2 can be obtained. Similarly, the variances S3, S4, and S5 corresponding to m=3-5 can be obtained.

[0043] 3. Based on the known data S, S1, S2, S3, S4, S5, and the variance data, determine whether the data after the initial startup is in a stable state. Based on the change in variance, the state of the current measurement environment can be determined. Assume that the upper limit of the variance stability value is set to S S1 =0.03, lower limit S S2 =0.02, the first step is to determine the total variance S and S S2 If S S2 , the data is stable. After receiving this feedback, the algorithm will reduce the cleaning frequency. The way to reduce the frequency is to increase the sampling period. For example, the buffer filling frequency is changed from one per minute to one per two minutes. At this time, the time required to fill the buffer cache will become 60 minutes. Similarly, the variance analysis of the buffer data is performed. If it continues for S S2 , continue to increase the sampling period by three minutes / s until S S2 ≤S≤S S1 , the cleaning brush activation frequency at this time is the cleaning frequency suitable for this environment. S1 , it means that the data is in an unstable state. At this time, it is necessary to analyze according to S1, S2, S3, S4, S5, respectively. First, a stable data S1, S2, S3 are all in S S1 and S​​S2 If the data is unstable due to the influence of the environment, then according to the way the buffer is iterated, S4>S S1 , at this time, the algorithm needs to determine whether there is a value greater than S in S1-S4 S1 Assuming that when the cleaning brush is running at an initial frequency of 30 minutes, S S2 ≤S1≤S S1 , S S2 ≤S2≤S S1 , S S2 ≤S3≤S S1 ,S4>S S1 ,S5>S S1 Taking the case as an example, the data instability anomaly can be located at the starting position of S4. When S3 indicates that m=3, the valid number in the buffer cache is m*6=3*6=18. The effective cleaning time T of the cleaning brush should be 18 minutes for cleaning. If it is not cleaned for more than 18 minutes, the data will be contaminated by environmental factors, resulting in data anomalies. This time is adjusted by adjusting the data sampling frequency. The sampling frequency = T / 30. Taking the current case as an example, T=18min, then f=18 / 30=0.6, and the sampling frequency of 0.6min / times is written to the sensor, the cleaning brush is started, and the current buffer is cleared for a new cache. When the cache is over, repeat step 2 until S S2 ≤S≤S S1 , the cleaning brush activation frequency at this time is the cleaning frequency suitable for this environment.

[0044] This adjustment method can effectively and dynamically adjust the start-up cycle of the cleaning brush to adapt to changes in the current environment.

[0045] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable ordinary technicians in the field to understand the content of the present invention and implement it. They cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made according to the spirit of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for adaptively adjusting the frequency of a sensor cleaning brush, characterized in that: include: Data collection step: taking the period between two consecutive cleaning starts of the cleaning brush as a cleaning cycle, and collecting sensor detection data for a preset number of times within the cleaning cycle; Data calculation and comparison steps: After data collection is completed, calculate the stability value of the full amount of data in this cleaning cycle and compare it with the preset stability standard value; when the actual measured stability value of the full amount of data is worse than the preset stability standard value, increase the cleaning frequency; when the actual measured stability value of the full amount of data is better than the preset stability standard value, reduce the cleaning frequency; To obtain the corrected cleaning frequency; when the actual stability value of the full amount of data is within the preset stability standard value range, the cleaning frequency remains unchanged; Thereby obtaining a corrected cleaning frequency value for correcting the next cleaning cycle; Cleaning cycle modification step: after the cleaning brush starts cleaning again, the duration of the new cleaning cycle is modified based on the modified cleaning frequency value obtained in the previous cleaning cycle; In the new cleaning cycle, the sensor detection data is collected for a preset number of times, and a new corrected cleaning frequency is calculated to modify the duration of the next cleaning cycle; In the data calculation and comparison step, the stability value of the full amount of data is calculated according to the variance formula to obtain the variance S of the full amount of data; The preset stability standard value is the upper limit of the variance S S1 , the lower limit of variance is S S2 ; When S< S S2 When S>S S1 When S S2 ≤S≤S S1 The cleaning frequency remains unchanged; In the data collection step, within the current cleaning cycle of duration T, the preset number of sensor detection data collection is K, and the K data are divided into m groups according to the order of collection time, each group contains n data, where K=m*n; In the data calculation and comparison step, when S>S S1 When the mean of K data is used to calculate the variance of each group of data in m groups, the variance of the first group of data is S1, the variance of the second group of data is S2, and so on until the variance of the mth group of data is S m , S 1、 S2 …S m Sequentially with the upper limit of variance S S1 For comparison, if the variance S of the tth group of data is t Start larger than S S1 , then the next cleaning cycle is modified to (t-1)*T / m, where 1≤t≤m.

2. The method for adaptively adjusting the frequency of a sensor cleaning brush according to claim 1, characterized in that: After the sensor is powered on, the cleaning brush starts cleaning for the first time, and the preset cleaning frequency is used to determine the duration of the cleaning cycle between the first cleaning and the second cleaning.

3. The method for adaptively adjusting the frequency of a sensor cleaning brush according to claim 1, characterized in that: In the data calculation and comparison step, when S S2 When the next cleaning cycle is changed to a*T, a is a constant, 1 <a≤2。​ 4. The method for adaptively adjusting the frequency of a sensor cleaning brush according to claim 3, characterized in that: The number of groups m of K data satisfies: [K 1 / 2 ]-1≤m≤[K 1 / 2 ]+1.

Images