A control method for stable power supply of sprinkler slip ring

Abstract

The present invention relates to the field of motor control technology, and specifically to a control method for a stable power supply of a sprinkler slip ring, the method collects flow data in real time to obtain the flow stability of each unit time period; obtains the characteristic vector of each unit time period and filters out the abnormal characteristic vector, and then obtains the abnormal threshold; uses the characteristic vector corresponding to the flow stability greater than the abnormal threshold as the first target vector, and classifies to obtain multiple state categories, and obtains the speed range corresponding to each state category; uses the characteristic vector corresponding to the flow stability not greater than the abnormal threshold as the second target vector, and obtains the outlier factor of each second target vector; and then obtains the conservative speed range of each second target vector; obtains the predicted conservative speed range and outlier factor, and then obtains the reliable speed, and regulates the speed. The present invention can avoid the problem of over-speed steering due to the loss of step phenomenon, and stabilizes the working state of the sprinkler slip ring power supply.

Description

Technical Field

[0001] The invention relates to the technical field of motor control, and in particular to a control method for stable power supply of a slip ring of a sprinkler. Background Art

[0002] During operation, sprinklers need to rotate continuously to irrigate large areas of farmland. Slip ring power supply control uses slip ring devices to establish electrical connections between the rotating part and the fixed part of the sprinkler. Slip rings usually consist of a rotating conductive ring and a set of fixed brushes. When the sprinkler rotates, the conductive ring rotates with it, while the brushes remain fixed, and power is transmitted through the contact between the two; among them, changes in the sprinkler flow rate will affect the load of the rotating part. When the sprinkler flow rate increases, the reaction force of the water flow sprayed by the sprinkler head on the rotating part will also increase, which may cause the speed of the rotating part to decrease. The change in the speed of the rotating part will change the movement trajectory and water spraying angle of the sprinkler head, thereby affecting the sprinkler flow rate. When the speed increases, the centrifugal force of the sprinkler head increases, which may change the water spraying angle, affect the speed stability of the slip ring part, and then cause pressure fluctuations between the carbon brush and the slip ring in the system, reducing the stability of the sprinkler slip ring power supply.

[0003] The speed feedback of the rotating part of the traditional sprinkler slip ring power supply is usually achieved by using Hall elements. However, in the case of high-speed motors, due to the large amount of data to be processed, the reaction time of the Hall element is relatively long, and it is difficult to accurately control the motor speed to the required value. This may cause the motor to turn too fast, which in turn has an adverse effect on the stability of the sprinkler slip ring power supply. Specifically, during the operation of the high-speed motor, the Hall element may not be able to detect the speed change in a timely and accurate manner, making it difficult for the control system to make quick adjustments, resulting in large fluctuations in the motor speed, affecting the stability and reliability of the sprinkler slip ring power supply. Summary of the invention

[0004] In order to solve the problem of low power supply stability caused by inaccurate speed control of the rotating part of the slip ring power supply in the sprinkler, the present invention provides a control method for stable power supply of the slip ring of the sprinkler, and the technical scheme adopted is as follows:

[0005] An embodiment of the present invention provides a control method for a sprinkler slip ring stable power supply, the method comprising the following steps:

[0006] Real-time detection of the flow velocity at the sprinkler outlet. In each unit period, the uniformity of the flow velocity change is obtained according to the change of the flow velocity at adjacent moments, and the flow velocity stability is obtained in combination with the real-time flow velocity data.

[0007] The rotation speed of the sprinkler slip ring and the pressure between the carbon brush and the slip ring are collected in real time, and the pressure stability is obtained according to the distribution difference of all pressures in a unit time period. The product of the uniformity of the flow rate change and the pressure stability is used as the sprinkler stability;

[0008] The corresponding feature vector is composed of the velocity data sequence, velocity stability and slip ring speed sequence corresponding to each unit time period, all feature vectors are judged abnormally, the abnormal feature vectors are screened out, and the abnormal threshold is obtained based on the sprinkler irrigation stability corresponding to the abnormal feature vector;

[0009] Taking the characteristic vector corresponding to the sprinkler irrigation stability greater than the abnormal threshold as the first target vector, obtaining the corresponding relative distance according to the similarity and time difference between every two first target vectors, classifying all the first target vectors based on the relative distance to obtain multiple state categories, and obtaining the speed range corresponding to each state category;

[0010] Taking the characteristic vector corresponding to the sprinkler irrigation stability not greater than the abnormal threshold as the second target vector, assigning the second target vector to the state category where the first target vector with the smallest relative distance thereto is located, and in each state category, obtaining the corresponding distance function based on the corresponding speed range and the speeds corresponding to every two second target vectors, thereby obtaining the outlier factor of each second target vector;

[0011] The conservative speed range of the second target vector is obtained according to the speed range corresponding to the state category of the second target vector and the corresponding outlier factor; the predicted conservative speed range and outlier factor are obtained, and then the reliable speed is obtained to regulate the speed of the sprinkler slip ring.

[0012] In some embodiments, the method for obtaining the flow rate stability is:

[0013] The ratio of the flow velocity data at every two adjacent moments is obtained, and the uniformity of the flow velocity change is obtained based on all ratios within a unit time period; the average value of all the flow velocity data within a unit time period is calculated, and multiplied by the uniformity of the flow velocity change to obtain the flow velocity stability.

[0014] In some embodiments, the method for obtaining the pressure stability is:

[0015] The standard deviation and range of all pressures in a unit time period are calculated, and the pressure stability is obtained based on the standard deviation and range. Both the standard deviation and the range are negatively correlated with the pressure stability.

[0016] In some embodiments, the screening out abnormal feature vectors includes:

[0017] By setting the sampling ratio, all feature vectors are judged by a single classifier to screen out an abnormal vector set; the intersection of the abnormal vector sets obtained by screening for a preset number of times is taken, and all feature vectors in the intersection are used as the screened abnormal feature vectors.

[0018] In some embodiments, the method for obtaining the abnormal threshold is:

[0019] The sprinkler irrigation stabilities corresponding to all abnormal feature vectors are sorted in order of size, and a first preset proportion of sprinkler irrigation stabilities are selected from small to large to calculate an average value as the abnormal threshold.

[0020] In some embodiments, the relative distance is obtained by:

[0021] For any two first target vectors, the cosine similarity between the two is calculated as the similarity, and then the difference between the two is obtained, and the sum of the similarity and the difference is a preset value; the time difference between the two in time sequence is obtained, and the relative distance is obtained in combination with the difference; the time difference is positively correlated with the relative distance.

[0022] In some embodiments, the method for obtaining the speed range is:

[0023] For each state category, the sprinkler slip ring speeds corresponding to the first target vectors are sorted by size, and the average value of the sprinkler slip ring speeds of the second preset proportion is selected from large to small as the upper limit of the speed range, and the average value of the sprinkler slip ring speeds of the second preset proportion is selected from small to large as the lower limit of the speed range to form the speed range.

[0024] In some embodiments, the distance function is obtained by:

[0025] For any two second target vectors in each state category, the relative distance between the two second target vectors is calculated; the average value of the rotational speeds corresponding to the two target vectors is obtained, and the average value of the rotational speed range is obtained, the absolute value of the difference between the two average values ​​plus a preset value is added to obtain an expansion coefficient, and the product of the relative distance between the two second target vectors and the corresponding expansion coefficient is the distance function of the two second target vectors.

[0026] In some embodiments, the method for obtaining the conservative speed range is:

[0027] All outlier factors are normalized. For each second target vector, the upper limit of the conservative speed range is obtained according to the average value, maximum value and normalized outlier factor of the speed range in the state category; the lower limit of the conservative speed range is obtained according to the average value, minimum value and normalized outlier factor of the speed range in the state category; and the conservative speed range corresponding to the second target vector is formed.

[0028] In some embodiments, the method for obtaining the reliable rotation speed is:

[0029] The product of the lower limit of the predicted conservative speed range and the predicted outlier factor is calculated, the product of the difference between the preset value and the predicted outlier factor and the upper limit of the predicted conservative speed range is calculated, and the two products are added to obtain the reliable speed.

[0030] The embodiments of the present invention have at least the following beneficial effects:

[0031] First, the data of the sprinkler are collected: flow rate data, sprinkler slip ring speed, and pressure between the carbon brush and the slip ring. Then, the flow rate stability and sprinkler irrigation stability are obtained through data integration. The flow rate stability characterizes the change of flow rate in the water outlet of the sprinkler irrigation machine in a unit period, and the sprinkler irrigation stability characterizes the change of the sprinkler irrigation state at the water outlet of the sprinkler irrigation machine in a unit period. Then, the features of each unit period extracted are represented by forming feature vectors, and the abnormal feature vectors are screened out to obtain the abnormal threshold. The abnormal threshold is used to evaluate whether the sprinkler irrigation stability is abnormal. The feature vector corresponding to the sprinkler irrigation stability greater than the abnormal threshold is used as the first target vector. At this time, the sprinkler irrigation stability corresponding to the unit period is higher, and the sprinkler irrigation state change at the water outlet is more stable. By classifying all the first target vectors, the unit periods with similar working conditions are divided into the same state category, and then the corresponding speed range is obtained as the speed change range under normal working conditions. Then, the characteristic vector corresponding to the sprinkler stability that is not greater than the abnormal threshold is used as the second target vector. At this time, the sprinkler stability corresponding to the unit time period is low, and the working state is abnormal. The second target vector is assigned to the nearest state category according to the principle of the closest relative distance, and then the outlier factor of the second target vector in each state category is calculated. The speed range of the state category is corrected by the outlier factor to obtain the conservative speed range corresponding to each second target vector, that is, for each unit time period with abnormal working state, there is a conservative speed range that takes the characteristic vector of the abnormal state into account. When the slip ring speed in the abnormal state is within the conservative speed range, it can make the sprinkler operation easier to synchronize, and then the slip ring power supply state is stable. Then the conservative speed range and outlier factor are predicted to obtain the reliable speed, and the sprinkler slip ring speed is regulated. According to the obtained reliable speed, the most suitable speed in the current state is obtained. At this speed, the sprinkler slip ring power supply working state is stable and easier to synchronize, which can avoid the problem of too fast steering caused by the loss of step phenomenon, so that the sprinkler slip ring power supply working state is stable and improves work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are 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.

[0033] Figure 1 A flowchart of a method for controlling a stable power supply of a sprinkler slip ring provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the control method of the slip ring stable power supply of a sprinkler proposed by the present invention, its specific implementation method, structure, features and effects in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0035] 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.

[0036] The specific scheme of the control method for stable power supply of a sprinkler slip ring provided by the present invention is described in detail below with reference to the accompanying drawings.

[0037] See also Figure 1 , which shows a flow chart of the steps of a control method for a stable power supply of a sprinkler slip ring provided by an embodiment of the present invention, the method comprising the following steps:

[0038] Step S001, real-time detection of the flow velocity at the water outlet of the sprinkler, in each unit time period, the uniformity of flow velocity change is obtained according to the change of flow velocity at adjacent moments, and the flow velocity stability is obtained in combination with the real-time flow velocity data.

[0039] The speed of the slip ring of the sprinkler acts on the rotation of the sprinkler head. When the sprinkler head rotates, the power supply slip ring rotates accordingly, the sprinkler head sprays the liquid, and the internal slip ring provides stable power supply. Therefore, the speed of the slip ring will affect the sprinkler irrigation state.

[0040] For sprinklers, since there may be stability differences between the rotation of the sprinkler head and the speed of the sprayed water during the sprinkler irrigation process, the flow rate of the water outlet should be measured first. Since the application environment of the sprinkler is farmland irrigation, the sprinkler can irrigate such areas by spraying water.

[0041] When the sprinkler is irrigating, if the speed change of the sprinkler slip ring leads to local carbonization and uneven pressure between the slip rings, which in turn reduces the stability of the slip ring power supply, it will increase the load on the sprinkler motor and may cause asynchronism.

[0042] There are many methods for real-time detection of the flow rate at the water outlet of the sprinkler. In the embodiment of the present invention, a flow rate sensor is installed at the water outlet of the sprinkler to collect real-time flow rate data. The data is collected once per second, and the flow rate data of the i-th second is recorded as .

[0043] Since the flow rate change at the water outlet may fluctuate due to interference in the application environment, it is necessary to calculate the change in the flow rate at the water outlet. Ideally, the water flow velocity at the water outlet is within a reasonable range of change.

[0044] Taking 10 seconds as a unit time period, the change of the outlet flow rate per unit time period is determined to characterize the outlet flow rate stability.

[0045] The ratio of the flow velocity data at every two adjacent moments is obtained, and the uniformity of flow velocity change is obtained based on all ratios within the unit time period; the average value of all flow velocity data within the unit time period is calculated, and multiplied by the uniformity of flow velocity change to obtain the flow velocity stability.

[0046] By calculating the ratio of the flow velocity data at adjacent moments within a unit period, the fluctuation characteristics of the water flow velocity within 10 seconds during the sprinkler irrigation process are characterized. The ratio of the data of the two seconds before and after is used to express the rate of change. The closer the rate of change is to 1, the smaller the flow velocity change is. Then, the absolute value of the difference between each ratio and 1 is obtained and the average value is calculated to represent the difference in the flow velocity change at the outlet within 10 seconds. The uniformity of the flow velocity change is obtained by subtracting the difference from 1. ,in, represents the flow rate data at the i-th second, represents the flow rate data of the i+1th second, represents the flow rate data of the i+8th second, Represents the flow velocity data of the i+9th second. The average value of all flow velocity data in a unit period is taken as the representative value of the flow velocity data of the unit period. The representativeness of the mean value is determined by the volatility of these ten flow velocity data. The higher the uniformity of the outlet flow velocity change, the better the representativeness of the mean value. Therefore, the mean value is used as the correction coefficient of the flow velocity uniformity to obtain the flow velocity stability of the unit period. Taking the jth unit period as an example, the flow velocity stability is calculated. : in, represents the flow rate data at the i-th second, represents the flow rate data at the i+9th second, represents the averaging function, Indicates the uniformity of flow velocity change in the jth unit time period.

[0047] The closer the ratio of the flow velocity data at adjacent moments is to 1, the smaller the change in flow velocity between two adjacent moments. Correspondingly, the smaller the average value of the absolute value of the difference between the ratio and 1 in a unit period, the more uniform the flow velocity change, indicating that the change in the outlet flow velocity in the unit period is smaller and the flow velocity stability is greater. When the degree of uniformity of flow velocity change is greater, the representativeness of the mean of flow velocity data is better, and the flow velocity stability obtained by combining the mean is greater.

[0048] Step S002, real-time acquisition of the rotation speed of the sprinkler slip ring and the pressure between the carbon brush and the slip ring, obtaining the pressure stability according to the distribution difference of all pressures in a unit time period, and taking the product of the uniformity of the flow rate change and the pressure stability as the sprinkler stability.

[0049] The process of sprinkler irrigation is that the power of the sprinkler increases and the sprinkler speed increases as needed, that is, the range of sprinkler irrigation gradually increases within a unit period of time. Since sprinkler irrigation is a continuous process, a stable slip ring power supply is required. However, during the application process, the water flow from the sprinkler head will generate a reaction force on the rotating part, and the water flow rate will fluctuate due to environmental interference, which will affect the speed of the sprinkler irrigation process and the stability of the pressure between the carbon brush and the slip ring. Therefore, the load borne by the sprinkler at different times is unknown and may vary.

[0050] Theoretically, when the flow rate is high, the controller should increase the output power of the sprinkler, and the output power is related to the speed. The higher the speed of the motor, the greater the pressure between the carbon brush and the slip ring of the sprinkler per unit time.

[0051] However, due to the existence of this unknown load interference, the controller can only use a more conservative load angle. Therefore, when the load increases, the speed of the sprinkler will first decrease, and then the controller can speed up the sprinkler based on the correlation between the speed and the load angle. In this way, the speed of the sprinkler is uncertain, that is, the speed of the slip ring is also uncertain; and the control method of the controller cannot match this multi-modal interference, so the pressure between the corresponding carbon brush and the slip ring will also decrease after the speed decreases. Not only that, it may also cause a sudden change in the pressure between the carbon brush and the slip ring, which in turn affects the operating state of the sprinkler.

[0052] The actual slip ring speed is collected by the speed sensor, and the sampling is synchronized with the flow rate data, that is, the real-time speed is collected once per second, and the real-time speed at the i-th second is recorded as .

[0053] In the actual production process, there will be water flow vibration in the water pipeline, that is, the vibration effect of the change of water flow velocity on the sprinkler. The water flow vibration is caused by the reaction force of the water flow in the sprinkler area caused by the change of flow rate and water flow velocity in the water pipeline. In severe cases, the rotation of the sprinkler is disordered or tilted. It can be inferred from the above that if the change of water flow state causes the reaction force on the sprinkler to change higher, it means that the speed of the sprinkler nozzle is disordered at this time. Even if the speed is high, the pressure between the carbon brush and the slip ring in the sprinkler will be disordered. In severe cases, it will cause the power supply of the slip ring to be interrupted. This is a mismatch between high power and high efficiency.

[0054] A pressure sensor is installed on the slip ring of the sprinkler. The pressure at the corresponding moment can be collected from the pressure sensor. The pressure at the i-th second is recorded as .

[0055] Since 10 pressure data are collected in a unit time period, the change in pressure reflects the severity of the impact of the reaction force change on the nozzle. Therefore, the pressure stability is calculated through data that can reflect the pressure change, namely the standard deviation and range. Specifically, the standard deviation and range of all pressures in the unit time period are calculated, and the pressure stability is obtained based on the standard deviation and range. Both the standard deviation and range are negatively correlated with the pressure stability. Taking the jth unit time period as an example, the specific calculation formula is: in, Indicates the pressure stability of the jth unit period; represents the pressure at the i-th moment, represents the pressure at the i+9th moment, represents the standard deviation of 10 pressure data in the jth unit period; e represents a natural constant; represents the maximum value function, represents the minimum function, It represents the difference between the maximum and minimum values ​​of the 10 pressure data in the j-th unit time period, that is, the range.

[0056] The 1 in the denominator of the above formula is to ensure that the denominator is not 0. The denominator represents the change amplitude of the pressure between the carbon brush and the slip ring within 10 seconds of the jth unit period in a range. When the pressure between the carbon brush and the slip ring remains stable within 10 seconds, it means that the pressure stability during this period is good; The standard deviation is used to characterize the stability of the pressure between the carbon brush and the slip ring within 10 seconds. The larger the value, the higher the stability. This means that although the sprinkler is affected by the change of the reaction force within 10 seconds, the slip ring of the sprinkler will not have unstable power supply due to the reaction force disorder. The load change borne by the sprinkler at this time is relatively stable.

[0057] The product of flow rate change uniformity and pressure stability is taken as the sprinkler irrigation stability. Taking the jth unit period as an example, the sprinkler irrigation stability is , It is to take the average of the changes in flow velocity data at adjacent moments, assuming that the spraying water velocity is constant in a short period of time. If there is a large change in the water flow reaction force, data with large fluctuations will be obtained during the measurement process. It is the pressure stability of the jth unit period, reflecting the stability of pressure. The change of the sprinkler state of the water outlet within 10 seconds is characterized by the product of the uniformity of flow rate change and pressure stability calculation, which further reflects the sprinkler stability within 10 seconds. The mismatch factor caused by the change of the nozzle reaction force caused by the vibration of the water flow in the water pipeline is considered, and the change of sprinkler stability is jointly constrained. The larger the value of sprinkler stability, the more stable the sprinkler state within 10 seconds.

[0058] Step S003, the velocity data sequence, velocity stability and slip ring speed sequence corresponding to each unit time period form a corresponding feature vector, all feature vectors are judged as abnormal, abnormal feature vectors are screened out, and the abnormal threshold is obtained based on the sprinkler irrigation stability corresponding to the abnormal feature vector.

[0059] When using a sprinkler for irrigation, the speed of the slip ring is mechanically driven to provide a stable power supply to the sprinkler. However, due to the reaction force during spraying and the actual power supply efficiency of the sprinkler, the control strategy is not optimal, and there are potential problems of low efficiency, high energy consumption, and high risk of control failure. Therefore, starting from the irrigation stability corresponding to the theoretical optimal speed, the optimal speed is found under this type of irrigation stability.

[0060] Therefore, the abnormal threshold of irrigation stability is first obtained for subsequent screening of irrigation stability.

[0061] The corresponding feature vector is composed of the flow rate data, flow rate stability and slip ring speed sequence corresponding to each unit time period. Taking the jth unit time period as an example, the corresponding feature vector ,in represents the velocity data sequence of the jth unit period, ; represents the flow velocity stability in the jth unit period; represents the slip ring speed sequence of the jth unit period, . Eigenvector There are 21 parameters in it, the first 10 parameters are ; then add parameters by interpolation ; The last 10 parameters are .

[0062] By setting the sampling ratio, all feature vectors are judged by a single classifier to screen out an abnormal vector set; the intersection of the abnormal vector sets obtained by screening for a preset number of times is taken, and all feature vectors in the intersection are used as the screened abnormal feature vectors.

[0063] In this embodiment, the sampling ratio is set to 70%, and a single classifier OCSVM is used to judge all feature vectors: all recorded data are input into OCSVM, so as to ensure that , and When the indicator is not common, OCSVM will identify it as an outlier in the data, i.e., an abnormal point. Based on this, the abnormal vector set is screened out from all feature vectors.

[0064] The single classifier OCSVM is used to repeatedly sample, train, and discriminate several times. As an example, in this embodiment, it is three times. Each time, an abnormal vector set is screened out. The three abnormal vector sets may not be exactly the same, so the intersection data of the three abnormal vector sets are obtained as the abnormal feature vector.

[0065] The sprinkler irrigation stabilities corresponding to all abnormal feature vectors are sorted in order of size, and a first preset proportion of sprinkler irrigation stabilities are selected from small to large to calculate an average value as the abnormal threshold.

[0066] As an example, in an embodiment of the present invention, the first preset ratio is 50%, and the sprinkler irrigation stabilities corresponding to all abnormal feature vectors are arranged from large to small, and the average value of the last 50% of the sprinkler irrigation stabilities is selected and calculated as the abnormal threshold.

[0067] Step S004, taking the characteristic vector corresponding to the sprinkler irrigation stability greater than the abnormal threshold as the first target vector, obtaining the corresponding relative distance according to the similarity and time difference between every two first target vectors, classifying all the first target vectors based on the relative distance to obtain multiple state categories, and obtaining the speed range corresponding to each state category.

[0068] The characteristic vector corresponding to the sprinkler irrigation stability greater than the abnormal threshold is used as the first target vector. The greater the sprinkler irrigation stability, the more stable the working state. The abnormal threshold is obtained by calculating the average value of the part with smaller sprinkler irrigation stability, that is, the sprinkler irrigation stability greater than the abnormal threshold is a normal working state. For any two first target vectors, the corresponding relative distance is obtained by calculating the similarity and time difference. Specifically, for any two first target vectors, the cosine similarity between the two is calculated as the similarity, and then the difference between the two is obtained. The sum of the similarity and the difference is the preset value; the time difference between the two in the time series is obtained, and the relative distance is obtained in combination with the difference; the time difference is positively correlated with the relative distance.

[0069] Take the pth first target vector and the qth first target vector Take this as an example to calculate the corresponding relative distance ,in, represents the pth first target vector and the qth first target vector The cosine similarity between ; represents the pth first target vector and the qth first target vector The time difference between the two corresponding unit time periods, the unit of the time difference is the unit time period, for example , indicating the pth first target vector and the qth first target vector The two corresponding unit periods differ by two unit periods; represents the minimum function, and 0.15 represents the correction value of Δt.

[0070] By selecting The minimum value among 3 takes into account the rotational inertia of the sprinkler. If the time difference is large, it is limited to 3. Otherwise, the value is taken according to the original time difference.

[0071] All unit time periods are sorted and classified according to the relative distance between every two first target vectors through the DBSCAN clustering algorithm to obtain multiple state categories. The unit time period corresponding to the first target vector in each state category has similar working states.

[0072] For each state category, the slip ring speeds corresponding to the first target vectors are sorted by size, and the average value of the slip ring speeds of the second preset ratio is selected from large to small as the upper limit of the speed range, and the average value of the slip ring speeds of the second preset ratio is selected from small to large as the lower limit of the speed range to form the speed range.

[0073] As an example, in the embodiment of the present invention, the second preset ratio is 20%, that is, for each state category, the slip ring speeds corresponding to the first target vectors are arranged in descending order, the first 20% of the slip ring speeds are selected to find the average value as the upper limit of the speed range, and the last 20% of the slip ring speeds are selected to find the average value as the lower limit of the speed range, to obtain the speed range consisting of the upper and lower limits of each state category.

[0074] Step S005, taking the characteristic vector corresponding to the sprinkler irrigation stability not greater than the abnormal threshold as the second target vector, assigning the second target vector to the state category where the first target vector with the smallest relative distance is located, in each state category, obtaining the corresponding distance function based on the corresponding speed range and the speeds corresponding to every two second target vectors, and then obtaining the outlier factor of each second target vector.

[0075] The characteristic vector corresponding to the sprinkler irrigation stability not greater than the abnormal threshold is used as the second target vector, that is, the characteristic vector corresponding to the abnormal working state. The relative distance between the second target vector and each first target vector is calculated, and the state category of the first target vector corresponding to the smallest relative distance is selected as the state category of the second target vector, so as to complete the purpose of assigning the second target vector to the state category of the first target vector with the smallest relative distance.

[0076] For any two second target vectors in each state category, the relative distance between the two second target vectors is calculated; the average value of the rotational speeds corresponding to the two target vectors is obtained, and the average value of the rotational speed range is obtained, the absolute value of the difference between the two average values ​​plus the preset value is added to obtain the expansion coefficient, and the product of the relative distance between the two second target vectors and the corresponding expansion coefficient is the distance function of the two second target vectors.

[0077] Based on the relative distance, the expansion coefficient is obtained by using the average value of the speed range corresponding to any two second target vectors and the speed range of the state category, which is used to correct the relative distance between the two second target vectors to obtain the distance function. Taking the mth second target vector and the nth second target vector as an example, the distance function between the two is calculated: in, represents the distance function between the mth second target vector and the nth second target vector, represents the relative distance between the mth second target vector and the nth second target vector, represents the speed corresponding to the mth second target vector, represents the speed corresponding to the nth second target vector, express and The average value of , Y represents the speed range of the state category where the mth second target vector and the nth second target vector are located, represents the average value of the speed range Y, Indicates the expansion factor.

[0078] Use the expansion coefficient to correct the difference in relative distance. and The average and the interval mean The greater the difference, the greater the value of the relative distance. The second target vector is the abnormal feature vector, and the rotation speed is unreasonable. The more unreasonable the rotation speed is, the distance to other samples will be corrected to a larger value when calculating the outlier factor in the later stage, thereby increasing the sensitivity to abnormal samples.

[0079] The above distance function is used as the distance when calculating the outlier factor, and the local outlier factor of each second objective function is calculated by the LOF algorithm, wherein the maximum K value of the K nearest neighbors when calculating the local outlier factor is set to 10 in this embodiment. The LOF algorithm is a prior art, and the embodiment of the present invention only replaces the straight-line distance between two points in the LOF algorithm with the distance function in the present invention, and improves the local outlier factor to obtain the outlier factor of each second objective function.

[0080] Step S006, obtaining a conservative speed range of the second target vector according to the speed range corresponding to the state category of the second target vector and the corresponding outlier factor; obtaining the predicted conservative speed range and outlier factor, and then obtaining a reliable speed to regulate the speed of the sprinkler slip ring.

[0081] All outlier factors are normalized. For each second target vector, the upper limit of the conservative speed range is obtained according to the average value, maximum value and normalized outlier factor of the speed range in the state category; the lower limit of the conservative speed range is obtained according to the average value, minimum value and normalized outlier factor of the speed range in the state category; and the conservative speed range corresponding to the second target vector is formed.

[0082] Each second objective function corresponds to an outlier factor, which represents the probability that the second objective function does not belong to the state category. All outlier factors are normalized by range standardization. At this time, the value of the outlier factor E=1 is the record of the worst case.

[0083] For each second target vector:

[0084] The upper limit of the conservative speed range is obtained according to the average value, maximum value and normalized outlier factor of the speed range of the state category: ; Obtain the lower limit of the conservative speed range based on the average value, minimum value and normalized outlier factor of the speed range of the state category: .Depend on and The conservative speed range that constitutes the second target vector .

[0085] Use the GRU neural network to obtain the predicted conservative speed range and outlier factor:

[0086] The conservative speed range of the second target vector is obtained according to different start-up and shutdown conditions. Therefore, when training the GRU neural network, each start-up and shutdown is regarded as a complete data sequence. The flow rate data sequence w, slip ring speed sequence p, and pressure sequence z corresponding to each unit time period of each feature vector in the data sequence constitute a three-channel sequence. All three-channel sequences are used as data sets. The label values ​​are the conservative speed range Z and the outlier factor E.

[0087] For the first target vector in the feature vector, the mark E is 0, and Z is the maximum and minimum values ​​of the speed range Y in the corresponding type; for the second target vector in the feature vector, E is the result calculated by the above method of calculating the outlier factor, and Z is Z calculated by the above process of calculating the conservative speed range.

[0088] In order to enable the GRU network to achieve predictive control in advance, the time between the label and the sample is advanced by 10 seconds, that is, one unit period. The loss function for training the GRU network is the MSE loss. The input sequence of three channels is w, p and z, and the output sequence of three channels is E, , .

[0089] For the result of a GRU network prediction, there are the following speed control strategies:

[0090] First, according to the output value of the GRU network, the reliable speed is obtained: the product of the lower limit of the predicted conservative speed range and the predicted outlier factor is calculated, the product of the difference between the preset value and the predicted outlier factor and the upper limit of the predicted conservative speed range is calculated, and the two products are added together to obtain the reliable speed. The specific calculation formula is: . Determine whether the current speed p exceeds the predicted reliable speed If it does not exceed the limit, the controller will continue to run according to its own control logic. Otherwise, the current speed will be limited to the reliable speed. .

[0091] This implementation uses the water flow velocity at the outlet and the pressure between the carbon brush and the slip ring as the basis to obtain the sprinkler irrigation stability at the corresponding moment, and uses every ten seconds as a unit time period, which is also the control interval, to perform real-time control on the sprinkler power supply to achieve the purpose of intelligent and efficient control, avoid too low a matching rate for the sprinkler power supply, and appropriately reduce the speed. Software control is the main way to explore the performance of the slip ring power supply and jointly optimize the reliability of the slip ring power supply, thereby automatically generating a feasible speed and power corresponding to the slip ring power supply of the sprinkler.

[0092] In summary, the embodiment of the present invention detects the flow velocity at the water outlet of the sprinkler in real time. In each unit time period, the uniformity of the flow velocity change is obtained according to the change of the flow velocity at adjacent moments, and the flow velocity stability is obtained in combination with the real-time flow velocity data; the rotation speed of the sprinkler slip ring and the pressure between the carbon brush and the slip ring are collected in real time, and the pressure stability is obtained according to the distribution difference of all pressures in the unit time period, and the product of the uniformity of the flow velocity change and the pressure stability is used as the sprinkler stability; the flow velocity data sequence, flow velocity stability and slip ring rotation speed sequence corresponding to each unit time period form a corresponding feature vector, all feature vectors are judged as abnormal, and the abnormal feature vectors are screened out, and the abnormal threshold is obtained based on the sprinkler stability corresponding to the abnormal feature vector; the feature vector corresponding to the sprinkler stability greater than the abnormal threshold is used as the first target vector, and the first target vector is calculated according to each two first target vectors. The corresponding relative distance is obtained based on the similarity and time difference between the quantities, and all the first target vectors are classified based on the relative distance to obtain multiple state categories, and the speed range corresponding to each state category is obtained; the characteristic vector corresponding to the sprinkler irrigation stability that is not greater than the abnormal threshold is used as the second target vector, and the second target vector is assigned to the state category where the first target vector with the smallest relative distance is located. In each state category, the corresponding distance function is obtained based on the corresponding speed range and the speed corresponding to each two second target vectors, and then the outlier factor of each second target vector is obtained; the conservative speed range of the second target vector is obtained according to the speed range corresponding to the state category where the second target vector is located and the corresponding outlier factor; the predicted conservative speed range and outlier factor are obtained, and then the reliable speed is obtained to regulate the speed of the sprinkler slip ring. The present invention can avoid the problem of over-speed steering due to the out-of-step phenomenon, and stabilize the working state of the sprinkler slip ring power supply.

[0093] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0094] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0095] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Modifications to the technical solutions recorded in the aforementioned embodiments, or equivalent replacement of some of the technical features therein, do not cause the essence of the corresponding technical solutions to deviate 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 control method for stable power supply of a sprinkler slip ring, characterized in that: The method comprises the following steps: Real-time detection of the flow velocity at the sprinkler outlet. In each unit period, the uniformity of the flow velocity change is obtained according to the change of the flow velocity at adjacent moments, and the flow velocity stability is obtained in combination with the real-time flow velocity data. The rotation speed of the sprinkler slip ring and the pressure between the carbon brush and the slip ring are collected in real time, and the pressure stability is obtained according to the distribution difference of all pressures in a unit time period. The product of the uniformity of the flow rate change and the pressure stability is used as the sprinkler stability; The corresponding feature vector is composed of the velocity data sequence, velocity stability and slip ring speed sequence corresponding to each unit time period, all feature vectors are judged abnormally, the abnormal feature vectors are screened out, and the abnormal threshold is obtained based on the sprinkler irrigation stability corresponding to the abnormal feature vector; Taking the characteristic vector corresponding to the sprinkler irrigation stability greater than the abnormal threshold as the first target vector, obtaining the corresponding relative distance according to the similarity and time difference between every two first target vectors, classifying all the first target vectors based on the relative distance to obtain multiple state categories, and obtaining the speed range corresponding to each state category; Taking the characteristic vector corresponding to the sprinkler irrigation stability not greater than the abnormal threshold as the second target vector, assigning the second target vector to the state category where the first target vector with the smallest relative distance thereto is located, and in each state category, obtaining the corresponding distance function based on the corresponding speed range and the speeds corresponding to every two second target vectors, thereby obtaining the outlier factor of each second target vector; The conservative speed range of the second target vector is obtained according to the speed range corresponding to the state category of the second target vector and the corresponding outlier factor; the predicted conservative speed range and outlier factor are obtained, and then the reliable speed is obtained to regulate the speed of the sprinkler slip ring.

2. A control method for stable power supply of a sprinkler slip ring according to claim 1, characterized in that: The method for obtaining the flow rate stability is: The ratio of the flow velocity data at every two adjacent moments is obtained, and the uniformity of the flow velocity change is obtained based on all ratios within a unit time period; the average value of all the flow velocity data within a unit time period is calculated, and multiplied by the uniformity of the flow velocity change to obtain the flow velocity stability.

3. A control method for stable power supply of a sprinkler slip ring according to claim 1, characterized in that: The method for obtaining the pressure stability is: The standard deviation and the range of all pressures in a unit time period are calculated, and the pressure stability is obtained based on the standard deviation and the range, and both the standard deviation and the range are negatively correlated with the pressure stability.

4. A control method for stable power supply of a sprinkler slip ring according to claim 1, characterized in that: The step of filtering out abnormal feature vectors includes: By setting the sampling ratio, all feature vectors are judged by a single classifier to screen out an abnormal vector set; the intersection of the abnormal vector sets obtained by screening for a preset number of times is taken, and all feature vectors in the intersection are used as the screened abnormal feature vectors.

5. The control method for stable power supply of a sprinkler slip ring according to claim 1, characterized in that: The method for obtaining the abnormal threshold is: The sprinkler irrigation stabilities corresponding to all abnormal feature vectors are sorted in order of size, and a first preset proportion of sprinkler irrigation stabilities are selected from small to large to calculate an average value as the abnormal threshold.

6. A control method for stable power supply of a sprinkler slip ring according to claim 1, characterized in that: The relative distance is obtained by: For any two first target vectors, the cosine similarity between the two is calculated as the similarity, and then the difference between the two is obtained, and the sum of the similarity and the difference is a preset value; the time difference between the two in time sequence is obtained, and the relative distance is obtained in combination with the difference; the time difference is positively correlated with the relative distance.

7. A control method for stable power supply of a sprinkler slip ring according to claim 1, characterized in that: The method for obtaining the speed range is: For each state category, the sprinkler slip ring speeds corresponding to the first target vectors are sorted by size, and the average value of the sprinkler slip ring speeds of the second preset proportion is selected from large to small as the upper limit of the speed range, and the average value of the sprinkler slip ring speeds of the second preset proportion is selected from small to large as the lower limit of the speed range to form the speed range.

8. The control method for stable power supply of a sprinkler slip ring according to claim 1, characterized in that: The method for obtaining the distance function is: For any two second target vectors in each state category, the relative distance between the two second target vectors is calculated; the average value of the rotational speeds corresponding to the two target vectors is obtained, and the average value of the rotational speed range is obtained, the absolute value of the difference between the two average values ​​plus a preset value is added to obtain an expansion coefficient, and the product of the relative distance between the two second target vectors and the corresponding expansion coefficient is the distance function of the two second target vectors.

9. A control method for stable power supply of a sprinkler slip ring according to claim 1, characterized in that: The method for obtaining the conservative speed range is: Normalize all outlier factors, and for each second target vector, obtain the upper limit of the conservative speed range according to the average value, maximum value and normalized outlier factor of the speed range of the state category; The lower limit of the conservative speed range is obtained according to the average value, minimum value and normalized outlier factor of the speed range of the state category; The conservative speed range corresponding to the second target vector is formed.

10. The control method for stable power supply of a sprinkler slip ring according to claim 1, characterized in that: The method for obtaining the reliable speed is: The product of the lower limit of the predicted conservative speed range and the predicted outlier factor is calculated, the product of the difference between the preset value and the predicted outlier factor and the upper limit of the predicted conservative speed range is calculated, and the two products are added to obtain the reliable speed.

Images