A kind of intelligent control method of sewage treatment equipment

Abstract

The present invention relates to the field of data processing technology, and more specifically, to an intelligent control method for sewage treatment equipment, including: real-time collection of water inlet flow data and energy consumption data of sewage treatment equipment; for each water inlet flow data point, determining the mutation coefficient of the water inlet flow data point; determining the filtering requirement of the water inlet flow data point; in response to the filtering requirement being not less than a preset filtering threshold, determining the number of adaptive filtering times of the water inlet flow data point; adaptive filtering of each water inlet flow data point through an NLM algorithm; and completing intelligent control of sewage treatment equipment based on the filtered water inlet flow data. The present invention adaptively obtains the number of filtering times of water inlet flow data points for filtering and denoising, retains abnormal data in the water inlet flow data, and more accurately identifies and filters out noise data therein, so that the sewage treatment equipment can be more accurately controlled intelligently through the existing control algorithm later.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more specifically, to an intelligent control method for sewage treatment equipment. Background Art

[0002] With the increasing amount of sewage discharged in industrial production and domestic water, water resources and ecological environment have been under great pressure. The regulations and standards for sewage treatment have become increasingly stringent, prompting enterprises to seek more effective solutions. The inlet flow data in the operation data of sewage treatment equipment has a vital impact on the normal use of sewage treatment equipment. Excessive inlet flow may cause the equipment to overload and fail to effectively treat sewage, thereby reducing the treatment efficiency; too small inlet flow may cause the equipment to fail to fully function, affecting the treatment quality of sewage. However, when using the existing control algorithm to intelligently control the inlet flow data of sewage treatment equipment, when collecting the inlet flow data of sewage treatment equipment, the electromagnetic interference generated by the surrounding equipment or other electrical equipment may affect the sensor reading, thereby generating noise data in the inlet flow data sequence, resulting in the reduction of the accuracy of the existing control algorithm for the intelligent control of the inlet flow data. Therefore, it is first necessary to perform data denoising on the collected inlet flow sequence to obtain the real inlet flow data, and then use the existing control algorithm to intelligently control the inlet flow data. The Non-Local Means Filtering Algorithm (NLM) algorithm takes into account the non-local similarity of data and can better retain the detailed features and mutation points of the data while denoising. For the inlet flow data, there may be some instantaneous flow changes. The NLM algorithm can avoid these changes from being over-smoothed to a certain extent when filtering and denoising the inlet flow data.

[0003] The patent document with the announcement number CN104462014B discloses a new method for filtering low-field nuclear magnetic resonance signals based on improved non-local mean, which belongs to the field of low-field nuclear magnetic resonance signal processing. The steps include: a. Inputting the low-field nuclear magnetic resonance CPMG echo signal to be denoised; b. Calculating the number of points between the first point of the input CPMG echo signal and the point where the amplitude of the echo signal drops to 30% of the first point amplitude, and using the number of points as the search window width of the entire filtering algorithm; c. Calculating the size of the similarity window according to the linear function based on the data variance of each point of the CPMG echo signal; d. Using the Stein unbiased risk estimation method to calculate the size of the optimal attenuation parameter of the non-local mean algorithm; e. Using the parameters obtained in steps b, c and d to perform final filtering on the signal according to the non-local mean algorithm.

[0004] However, the above patent document does not make improvements to the inlet flow data, and does not solve the problem that when using the NLM algorithm to denoise the inlet flow data, the noise data of the inlet flow usually has a similar numerical performance to the actual abnormal inlet flow data. At the same time, the NLM algorithm has the same number of filtering times for all data points, and the fixed number of filtering times will cause some actual abnormal inlet flow data to be smoothed out, thereby affecting the authenticity and integrity of the inlet flow data. At the same time, the fixed number of filtering times will also make it impossible for some noise data to be completely smoothed out, which will also affect the authenticity of the inlet flow data, thereby reducing the accuracy of the subsequent existing control algorithms for the intelligent control of sewage treatment equipment. Summary of the invention

[0005] In order to solve the problem that the NLM algorithm uses the same number of filtering times for all data points, affecting the authenticity and integrity of the water flow data, and thus reducing the accuracy of the subsequent existing control algorithm for intelligent control of sewage treatment equipment, the present invention proposes a sewage treatment equipment intelligent control method, which includes the following steps:

[0006] The water inlet flow data and energy consumption data of the sewage treatment equipment are collected in real time; for each water inlet flow data point, the mutation coefficient of the water inlet flow data point is determined according to the difference in the mean value of the reference data segment preset for the water inlet flow data point before and after the water inlet flow data point is removed, and the proportion of extreme points in the water inlet flow data point in the reference data segment, and the mutation coefficient is positively correlated with the difference in the mean value and the proportion of extreme points; the water inlet flow data point and the energy consumption data point at the corresponding time are determined according to the difference in the relative sizes of the water inlet flow data point and its corresponding time in the preset reference data segment to which they belong, and the mutation coefficient of the water inlet flow data point. The filtering requirement of the point is positively correlated with the difference in relative sizes and the mutation coefficient; in response to the filtering requirement being not less than a preset filtering threshold, the adaptive filtering times of the inlet flow data point are determined according to the filtering requirement of the inlet flow data point and the preset initial filtering times, and the adaptive filtering times are positively correlated with the filtering requirement and the preset initial filtering times; according to the adaptive filtering times of each inlet flow data point, each inlet flow data point is adaptively filtered by the NLM algorithm to obtain filtered inlet flow data; based on the filtered inlet flow data, the intelligent control of the sewage treatment equipment is completed.

[0007] By dynamically adjusting the number of filters according to the mutation coefficient of each inlet flow data point, the noise caused by data fluctuations can be effectively eliminated, ensuring that the inlet flow data has higher authenticity and integrity, and improving data accuracy; by adaptively selecting the number of filters, more frequent filtering is ensured when data fluctuations are large, and unnecessary processing is reduced when the data is stable, thereby improving efficiency; accurate inlet flow data provides a reliable foundation for subsequent control algorithms, thereby improving the accuracy of intelligent control of sewage treatment equipment, helping to improve the efficiency and effectiveness of sewage treatment and ensure the stable operation of sewage treatment facilities.

[0008] Furthermore, the mutation coefficient satisfies the following relationship:

[0009] ; In the formula, For the The mutation coefficient of the inlet flow data point is For the The mean value of the reference data segment preset for each inlet flow data point, For the Remove the first water inlet flow data point from the preset reference data segment The mean value after the water inlet flow data points, Indicates The number of extreme value points of the inlet flow data points in the reference data segment preset for the inlet flow data points, Indicates The number of inlet flow data points in the reference data segment preset by the inlet flow data points. is a hyperparameter.

[0010] By comparing the difference between the current data point and the mean of the reference data segment after removing the current data point, potential mutations can be identified, which helps to detect abnormal traffic changes in a timely manner; by introducing the proportion of extreme points, the impact of a single outlier on the mean can be effectively suppressed, thereby improving stability and reliability in the presence of noise or extreme values.

[0011] Furthermore, the filtering requirement satisfies the following relationship:

[0012] ; In the formula, For the The filtering requirement of each inlet flow data point is For the The mutation coefficient of the inlet flow data point is: and Respectively The relative size of the water inlet flow data point and the energy consumption data point at the corresponding time in their respective preset reference data segments, is a hyperparameter, is the normalization function.

[0013] By combining the mutation coefficient and relative size, the filtering requirement is adjusted in real time to improve the accuracy of the inlet flow data; the filtering requirement is standardized through the normalization function to ensure that a consistent processing strategy can be maintained within different inlet flow data ranges, thereby improving the consistency and stability of data processing.

[0014] Furthermore, the relative size satisfies the following relationship:

[0015] , ; In the formula, and Respectively The relative size of the water inlet flow data point and the energy consumption data point at the corresponding time in their respective preset reference data segments, For the The value of the water inlet flow data point, For the The value of the energy consumption data point at the corresponding moment of the water inlet flow data point, and Respectively The maximum and minimum values ​​of each inlet flow data point in the reference data segment preset for each inlet flow data point, and Respectively The maximum and minimum values ​​of the energy consumption data points in the preset reference data segment of the energy consumption data point at the corresponding moment of the water inlet flow data point.

[0016] Furthermore, the number of adaptive filtering times satisfies the following relationship:

[0017] ; In the formula, For the The number of adaptive filtering times for each inlet flow data point, For the The filtering requirement of each inlet flow data point is is the preset filtering threshold, Indicates the preset initial filtering times, Is the rounding symbol.

[0018] By comparing the filtering requirement and the threshold, we can avoid mis-filtering normal data fluctuations and maintain the authenticity of the data. The number of adaptive filtering times is dynamically calculated based on the filtering requirement, and the intensity of the filtering can be automatically adjusted according to the changes in the actual data.

[0019] Furthermore, the adaptive filtering of each water inlet flow data point by the NLM algorithm comprises: in each filtering and denoising process of the NLM algorithm, searching in the water inlet flow data for a plurality of data segments with the smallest mean difference from the search data segments preset for the water inlet flow data point as neighborhood data segments, obtaining the weights of the data segments in each field according to the collection interval and mean difference between each neighborhood data segment and the search data segment, performing weighted summation of the means of all neighborhood data segments according to the weights, and assigning the result of the weighted summation to the water inlet flow data point to complete this filtering and denoising, and performing filtering and denoising for a corresponding number of times according to the number of adaptive filtering times of the water inlet flow data point.

[0020] Filtering is performed by finding neighborhood data segments that are similar to the preset search data segments of the target data points, so as to better retain the true characteristics of the water inlet flow data and reduce the loss of effective information; weights are calculated based on the mean difference and the acquisition interval to ensure that more similar neighborhood data segments contribute more to the weighted average, thereby more effectively removing noise and improving data accuracy; adaptive filtering based on the characteristics of data points can adapt to different data patterns and noise levels, thereby improving the flexibility and applicability of filtering.

[0021] Further, ; In the formula, For the The first search data segment preset for the inlet flow data point The weight of the neighborhood data segment, For the The search data segment preset for the inlet flow data point and its The collection interval between neighboring data segments is For the The search data segment preset for the inlet flow data point and its The mean difference between the neighboring data segments, For the The number of neighborhood data segments of the preset search data segment for the inlet flow data point.

[0022] Furthermore, the intelligent control of the sewage treatment equipment is achieved by using a PID control algorithm.

[0023] The present invention has the following beneficial effects:

[0024] In the process of filtering and denoising the water inlet flow data using the conventional NLM algorithm, some abnormal water inlet flow data will be erroneously smoothed by a fixed number of filtering times, while the noise data with a lower noise level will be ignored, resulting in inaccurate filtering. The present invention divides the water inlet flow data into segments and analyzes the filtering requirement according to the characteristics of the reference data segment, and adaptively obtains the number of filtering times for each water inlet flow data point according to the filtering requirement, thereby performing filtering and denoising. In this way, while retaining the abnormal data in the water inlet flow data, it is possible to more accurately identify and filter out noise data of various degrees, so that when the water inlet of the sewage treatment equipment is subsequently controlled by the existing control algorithm, the water inlet amount of the sewage treatment equipment can be more accurately controlled. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0026] Figure 1 It is a flowchart of the steps of an intelligent control method for sewage treatment equipment according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] 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 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0028] This method analyzes the changing characteristics of the water inlet flow data and energy consumption data of the sewage treatment equipment, obtains the filtering requirement of each water inlet flow data point, and then adaptively performs the corresponding filtering according to the filtering requirement of each water inlet flow data point, thereby achieving more complete data denoising and ensuring more accurate intelligent control of subsequent water inlet flow data.

[0029] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0030] See also Figure 1 , which shows a flowchart of a method for intelligent control of sewage treatment equipment provided by an embodiment of the present invention, the method comprising the following steps:

[0031] S1: Real-time collection of water flow data and energy consumption data of sewage treatment equipment.

[0032] The water flow data and energy consumption data of the sewage treatment equipment are collected in real time through flow sensors and smart meters, and then the analog-to-digital conversion equipment is used to digitize the above two types of data for subsequent analysis.

[0033] Implementers can set the collection frequency according to specific implementation conditions, for example, 2s / time.

[0034] S2: For each inlet flow rate data point, determine the mutation coefficient of the inlet flow rate data point.

[0035] It should be noted that this step is to analyze the numerical variation characteristics of the water inlet flow data to obtain the mutation coefficient of each water inlet flow data point; when analyzing this indicator, the more abnormal the numerical performance of each water inlet flow data point is, and the more abnormal the numerical performance of the data points around each water inlet flow data point is, the more likely it is that this water inlet flow data point belongs to noise data, and its mutation coefficient will be greater.

[0036] For each water inlet flow data point, the mutation coefficient of the water inlet flow data point is determined based on the difference in the mean values ​​of the reference data segment preset for the water inlet flow data point before and after the water inlet flow data point is removed, and the proportion of extreme value points in the water inlet flow data point in the reference data segment.

[0037] Specifically, the mutation coefficient satisfies the following relationship:

[0038] ;

[0039] In the formula, For the The mutation coefficient of the inlet flow data point is: For the The mean value of the reference data segment preset for each inlet flow data point, For the Remove the first water inlet flow data point from the preset reference data segment The mean value after the water inlet flow data points, Indicates The number of extreme value points of the inlet flow data points in the reference data segment preset for the inlet flow data points, Indicates The number of inlet flow data points in the reference data segment preset by the inlet flow data points. is a hyperparameter.

[0040] Implementers can set the length and hyperparameters of the reference data segment according to the specific implementation situation. For example, the length of the reference data point is centered on the inlet flow data point, with 20 inlet flow data points on each side. If there are less than 20 on either side, they will be supplemented on the other side, eventually reaching 41 inlet flow data points. Hyperparameters is 0.001, the hyperparameter The existence of is to prevent the situation where the mean values ​​of the reference data segment preset for the inlet flow data point are equal before and after the inlet flow data point is removed.

[0041] No. The difference between the mean values ​​before and after the water inlet flow data point. The larger the value, the The existence of the water flow data point makes the mean value of the entire reference data segment change more, then the The greater the possibility that the water inflow data point is abnormal, the The more likely a water inflow flow data point is to be noise data, the greater its mutation coefficient will be. Indicates The ratio of the number of extreme value points in the reference data segment preset for the inlet flow data point to the total number of data points in the reference data segment. The larger the value, the higher the ratio. The more frequently the data in the reference data segment of the inlet flow data point changes, the greater the possibility of noise data in the data segment. The more likely that an inlet flow data point belongs to noise data, the greater its credibility will be, and its mutation coefficient will also be greater.

[0042] S3: Determine the filtering requirement of the water inlet flow data point.

[0043] It should be noted that, according to the scenario survey, there is a certain correlation between the energy consumption data of the sewage treatment equipment and the water inlet flow data. This is because in the sewage treatment process, the equipment needs to consume more energy to handle higher flow. As the flow increases, the workload of the pump increases, resulting in an increase in energy consumption. At the same time, the power demand of the aeration system will also increase to ensure sufficient oxygen supply. Therefore, when the flow increases, the energy consumption usually increases accordingly. Therefore, in this step, the numerical change characteristics between the water inlet flow data and the energy consumption data will be analyzed, and the correlation between the two will be analyzed. Then, the filtering requirement of each water inlet flow data point is calculated in combination with the mutation coefficient of each water inlet flow data point. When analyzing the filtering requirement of each water inlet flow data point, the weaker the correlation between each water inlet flow data point and its corresponding energy consumption data, the greater the possibility that this water inlet flow data point belongs to noise data rather than the real data of the system, and the greater the corresponding filtering requirement. According to the above logic, the degree of demand for filtering times for each water inlet flow data point is calculated.

[0044] The filtering requirement of the water inlet flow data point is determined according to the difference in relative sizes of the water inlet flow data point and the energy consumption data point at the corresponding moment in their respective preset reference data segments, and the mutation coefficient of the water inlet flow data point.

[0045] Specifically, the filtering requirement satisfies the following relationship:

[0046] ;

[0047] In the formula, For the The filtering requirement of each inlet flow data point is For the The mutation coefficient of the inlet flow data point is: and Respectively The relative size of the water inlet flow data point and the energy consumption data point at the corresponding time in their respective preset reference data segments, is a hyperparameter, is the normalization function.

[0048] Implementers can set hyperparameters based on specific implementation conditions , for example, 0.001, hyperparameter The existence of is to prevent the situation that the water inlet flow data point and the energy consumption data point at the corresponding moment are respectively equal in relative size within their respective preset reference data segments.

[0049] in, The larger the The larger the mutation coefficient of the inflow flow data point is, the greater the possibility that it is a noise data point is. The greater the filtering requirement for each inlet flow data point; Indicates The relative difference between the water inlet flow data point and the energy consumption data point at the corresponding time in their respective preset reference data segments. The larger the value, the greater the difference between the two data points. The weaker the correlation between the sampling times corresponding to the inlet flow data points, the The greater the possibility that a water inlet flow data point belongs to noise data rather than real system data, the greater the corresponding filtering requirement will be.

[0050] Specifically, the relative size satisfies the following relationship:

[0051] , ;

[0052] In the formula, and Respectively The relative size of the water inlet flow data point and the energy consumption data point at the corresponding time in their respective preset reference data segments, For the The value of the water inlet flow data point, For the The value of the energy consumption data point at the corresponding moment of the water inlet flow data point, and Respectively The maximum and minimum values ​​of each inlet flow data point in the reference data segment preset for each inlet flow data point, and Respectively The maximum and minimum values ​​of the energy consumption data points in the preset reference data segment of the energy consumption data point at the corresponding moment of the water inlet flow data point.

[0053] S4: In response to the filtering requirement being not less than a preset filtering threshold, determining the number of adaptive filtering times for the water inlet flow data point.

[0054] It should be noted that through the above analysis, the filtering requirement of each water inlet flow data has been obtained. By setting the filtering threshold, the filtering requirement of each water inlet flow data point is If it is less than the filtering threshold, it can be said that the possibility of this water inlet flow data point being a noise data point is small, and it does not need to be filtered, that is, the number of filtering times is 0. If it is not less than the filtering threshold, it can be said that the possibility of this water inlet flow data point being a noise data point is large, and further operations are required to determine its filtering times.

[0055] The number of adaptive filtering times for the water inlet flow data point is determined according to the filtering requirement of the water inlet flow data point and a preset initial filtering number.

[0056] Specifically, the number of adaptive filtering times satisfies the following relationship:

[0057] ;

[0058] In the formula, For the The number of adaptive filtering times for each inlet flow data point, For the The filtering requirement of each inlet flow data point is is the preset filtering threshold, Indicates the preset initial filtering times, Is the rounding symbol.

[0059] Implementers can set the filtering threshold and the initial filtering times according to the specific implementation situation. For example, the filtering threshold is 0.2 and the initial filtering times is 5 times.

[0060] S5: Adaptively filter each water inlet flow data point using the NLM algorithm.

[0061] According to the number of adaptive filtering times of each water inlet flow data point, each water inlet flow data point is adaptively filtered by the NLM algorithm to obtain filtered water inlet flow data.

[0062] Specifically, the adaptive filtering of each water inlet flow data point by the NLM algorithm includes:

[0063] In each filtering and denoising process of the NLM algorithm, multiple data segments with the smallest mean difference from the search data segments preset for the inlet flow data point are searched in the inlet flow data as neighborhood data segments, and the weights of the data segments in each field are obtained according to the collection interval and the mean difference between each neighborhood data segment and the search data segment. The means of all neighborhood data segments are weighted and summed according to the weights, and the result of the weighted summation is assigned to the inlet flow data point to complete this filtering and denoising. According to the number of adaptive filtering times of the inlet flow data point, the corresponding number of filtering and denoising is performed.

[0064] Implementers can set the number of search data segments and multiple data segments according to specific implementation situations. For example, the search data segment is centered on the water inlet flow data point, with 10 water inlet flow data points on each side. If there are less than 10 data points on either side, they will be supplemented on the other side, eventually reaching 21 water inlet flow data points; the number of multiple data segments is the first 5 data segments with the smallest differences.

[0065] Specifically, the weight satisfies the following relationship:

[0066] ;

[0067] In the formula, For the The first search data segment preset for the inlet flow data point The weight of the neighborhood data segment, For the The search data segment preset for the inlet flow data point and its The collection interval between neighboring data segments is For the The search data segment preset for the inlet flow data point and its The mean difference between the neighboring data segments, For the The number of neighborhood data segments of the preset search data segment for the inlet flow data point.

[0068] Among them, the weight of the neighboring data segment with a smaller collection interval with the search data segment is greater, and the weight of the neighboring data segment with a smaller corresponding mean difference between data segments is greater. Indicates The weight factor when weighting the neighborhood data segments is: Represents the sum of weight factors when all neighborhood data segments are weighted.

[0069] S6: Based on the filtered water flow data, complete the intelligent control of the sewage treatment equipment.

[0070] Specifically, the intelligent control of the sewage treatment equipment is completed by using a PID control algorithm.

[0071] The implementer can set the proportional gain, integral gain and differential gain of the PID control algorithm according to the specific implementation situation. For example, the proportional gain is 0.6, the integral gain is 0.6 and the differential gain is 1.

[0072] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A sewage treatment equipment intelligent control method, characterized in that: include: Collect water flow data and energy consumption data of sewage treatment equipment in real time; For each water inlet flow data point, the water inlet flow data point and the water inlet flow data points around it are selected, and the reference data segment is preset. According to the difference between the mean values ​​of the water inlet flow data points in the reference data segment before and after the water inlet flow data point is removed, and the proportion of the number of extreme value points in the reference data segment to the total number of water inlet flow data points in the reference data segment, the mutation coefficient of the water inlet flow data point is determined, and the mutation coefficient is positively correlated with the difference in the mean value and the proportion of the extreme value points; according to the difference in relative sizes of the water inlet flow data point and the energy consumption data points at the corresponding time in their respective preset reference data segments, and the mutation coefficient of the water inlet flow data point, the filtering requirement of the water inlet flow data point is determined, and the filtering requirement is positively correlated with the difference in relative sizes and the mutation coefficient; In response to the filtering requirement being not less than a preset filtering threshold, determining the number of adaptive filtering times of the water inlet flow data point according to the filtering requirement of the water inlet flow data point and a preset initial filtering number, wherein the adaptive filtering number is positively correlated with the filtering requirement and the preset initial filtering number; According to the number of adaptive filtering of each water inlet flow data point, each water inlet flow data point is adaptively filtered by the NLM algorithm to obtain filtered water inlet flow data; Based on the filtered water flow data, intelligent control of sewage treatment equipment is completed.

2. The intelligent control method for sewage treatment equipment according to claim 1 is characterized in that: The mutation coefficient satisfies the following relationship: ； In the formula, For the The mutation coefficient of the inlet flow data point is: For the The mean value of the reference data segment preset for each inlet flow data point, For the Remove the first water inlet flow data point from the preset reference data segment The mean value after the water inlet flow data points, Indicates The number of extreme value points of the inlet flow data points in the reference data segment preset for the inlet flow data points, Indicates The number of inlet flow data points in the reference data segment preset by the inlet flow data points. is a hyperparameter.

3. The intelligent control method for sewage treatment equipment according to claim 1 is characterized in that: The filtering requirement satisfies the following relationship: ； In the formula, For the The filtering requirement of each inlet flow data point is For the The mutation coefficient of the inlet flow data point is: and Respectively The relative size of the water inlet flow data point and the energy consumption data point at the corresponding time in their respective preset reference data segments, is a hyperparameter, is the normalization function.

4. A sewage treatment equipment intelligent control method according to claim 1 or 3, characterized in that: The relative sizes satisfy the following relationship: ， ； In the formula, and Respectively The relative size of the water inlet flow data point and the energy consumption data point at the corresponding time in their respective preset reference data segments, For the The value of the water inlet flow data point, For the The value of the energy consumption data point at the corresponding moment of the water inlet flow data point, and Respectively The maximum and minimum values ​​of each inlet flow data point in the reference data segment preset for each inlet flow data point, and Respectively The maximum and minimum values ​​of the energy consumption data points in the preset reference data segment of the energy consumption data point at the corresponding moment of the water inlet flow data point.

5. The intelligent control method for sewage treatment equipment according to claim 1 is characterized in that: The adaptive filtering times satisfy the following relationship: ； In the formula, For the The number of adaptive filtering times for each inlet flow data point, For the The filtering requirement of each inlet flow data point is is the preset filtering threshold, Indicates the preset initial filtering times, Is the rounding symbol.

6. The intelligent control method for sewage treatment equipment according to claim 1 is characterized in that: The method of adaptively filtering each water inlet flow data point by using the NLM algorithm includes: In each filtering and denoising process of the NLM algorithm, multiple data segments with the smallest mean difference from the search data segments preset for the inlet flow data point are searched in the inlet flow data as neighborhood data segments, and the weights of the data segments in each field are obtained according to the collection interval and the mean difference between each neighborhood data segment and the search data segment. The means of all neighborhood data segments are weighted and summed according to the weights, and the result of the weighted summation is assigned to the inlet flow data point to complete this filtering and denoising. According to the number of adaptive filtering times of the inlet flow data point, the corresponding number of filtering and denoising is performed.

7. A sewage treatment equipment intelligent control method according to claim 6, characterized in that: The weights satisfy the following relationship: ； In the formula, For the The first search data segment preset for the inlet flow data point The weight of the neighborhood data segment, For the The search data segment preset for the inlet flow data point and its The collection interval between neighboring data segments is For the The search data segment preset for the inlet flow data point and its The mean difference between the neighboring data segments, For the The number of neighborhood data segments of the preset search data segment for the inlet flow data point.

8. The intelligent control method for sewage treatment equipment according to claim 1 is characterized in that: The intelligent control of the sewage treatment equipment is achieved by using a PID control algorithm.

Images