A method and system for harmless treatment of enterprise biochemical sludge

Abstract

This application relates to the field of sludge treatment technology, specifically to a method and system for the harmless treatment of enterprise biochemical sludge. The method includes: collecting hyperspectral images of the thickening tank of a gravity thickener at various times after the addition of flocculant; analyzing the floating of flocs on the liquid surface in the thickening tank and the changes in floc suspension caused by stirring during the thickening tank using real-time collected hyperspectral images; constructing the sludge flocculation coefficient and floc formation effect coefficient at each time point; combining the two to determine the appropriateness of flocculant addition at the current time; judging the appropriateness of the current flocculant dosage; and collecting the flocs in the thickening tank after sufficient flocculant has been added. This method solves the problem that the flocculant dosage cannot be adjusted in a timely manner according to the actual situation of the sludge during the current traditional sludge thickening process.

Description

Technical Field

[0001] This application relates to the field of sludge treatment technology, specifically to a method and system for the harmless treatment of enterprise biochemical sludge. Background Technology

[0002] Conventional sludge dewatering processes mainly include gravity thickening, centrifugal thickening, and mechanical dewatering. Among these, gravity thickening is currently the primary method for sludge treatment due to its low cost and mature technology. Gravity thickening is essentially a sedimentation process. Flocculants are added to and stirred in a first thickening tank, causing the sludge-water mixture to gradually form flocs that settle into a second thickening tank. A scraper continuously scrapes up the sludge flocs from the bottom of the second thickening tank, further thickening the sludge. Finally, an ultrasonic level gauge measures the sludge content in the second thickening tank. Once the sludge level in the second thickening tank reaches a threshold, a sludge pump is activated to remove the sludge, completing the sludge thickening and collection process.

[0003] In the process of treating sludge using gravity thickening, due to the variety of sludge types, the diverse components contained within the sludge, and the varying concentrations, flocculants are typically added at the initial stage of thickening. These flocculants aggregate fine particles in the sludge into larger flocs, thereby improving the settling speed and dewatering effect. However, in actual treatment, due to the varying concentrations and complex compositions of the sludge, the amount of flocculant added cannot be adjusted in a timely manner according to the actual situation. This can lead to insufficient addition, causing the sludge to remain suspended and float instead of settling, resulting in a slow settling rate and fewer flocs, which may also float and be discharged with the water flow, thus affecting subsequent treatment processes. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide a method and system for the harmless treatment of enterprise biochemical sludge, the specific technical solution of which is as follows:

[0005] In a first aspect, embodiments of this application provide a method for the harmless treatment of enterprise biochemical sludge, the method comprising the following steps:

[0006] Hyperspectral image data of the liquid level in the sludge thickening tank at various times after the addition of flocculant were collected;

[0007] Based on the similarity between the hyperspectral data of each pixel in the hyperspectral image at each time and the standard hyperspectral data, the flocculent pixels in the hyperspectral image at each time are determined; all flocculent pixels in the hyperspectral image at each time are clustered, and the sludge flocculation coefficient at the current time is calculated based on the number of clusters at the current time and the distance between clusters, combined with the difference between the clustering results at the current time and the first time.

[0008] Based on the trend of similarity between the hyperspectral image data of all previous times and the preset standard hyperspectral image data, and combined with the degree of disorder in the distribution of the number of clusters of all previous times, the floc formation effect coefficient of the current time is calculated.

[0009] The appropriateness of flocculant addition at the current moment is determined based on the sludge flocculation coefficient and the floc formation effect coefficient, so as to determine whether flocculant needs to be added at the current moment; after sufficient flocculant is added, the flocs in the thickening tank are settled and collected.

[0010] In one embodiment, the process of acquiring the flocculent pixels is as follows:

[0011] Pixels whose similarity to the hyperspectral data is greater than or equal to the preset similarity threshold are designated as floc pixels.

[0012] In one embodiment, the process of obtaining the sludge flocculation coefficient is as follows:

[0013] Calculate the variance of the distance between any two clusters at the current time; calculate the mean of the number of pixels in all clusters at each time; calculate the difference between the mean at the current time and the mean at the first time, and denote it as the first difference;

[0014] The sludge flocculation coefficient at the current moment is positively correlated with the number of clusters and the variance at the current moment, and negatively correlated with the first difference.

[0015] In one embodiment, the distance between the clusters is the distance between the cluster center points.

[0016] In one embodiment, the sludge flocculation coefficient at the current moment is the result of the first difference calculated by multiplying the product of the number of clusters at the current moment and the variance.

[0017] In one embodiment, the process of obtaining the floc formation effect coefficient is as follows:

[0018] The hyperspectral data sequence of each hyperspectral image data is determined by the hyperspectral data of all pixels in each hyperspectral image data; the similarity of the hyperspectral data sequence between the hyperspectral image data at the current time and the preset standard hyperspectral image data is calculated as the reactivity at the current time; the difference between the reactivity at two adjacent times is calculated and recorded as the second difference.

[0019] The trend strength is obtained by using the sequence of responsiveness from all previous moments as input to the trend algorithm.

[0020] Calculate the variance of the number of clusters at all times before the current time, and denote it as the first variance;

[0021] The floc formation effect coefficient at the current moment is positively correlated with the second difference and the trend intensity, and negatively correlated with the first variance.

[0022] In one embodiment, the process of acquiring the hyperspectral data sequence is as follows:

[0023] The fusion result of the same band data of all pixels in the hyperspectral image is used as the value of the corresponding band of the hyperspectral image. The values ​​of all bands of the hyperspectral image are obtained to construct the hyperspectral data sequence of the hyperspectral image.

[0024] In one embodiment, the appropriateness of flocculant addition is the normalized value of the product of the sludge flocculation coefficient and the floc formation effect coefficient.

[0025] In one embodiment, the process of determining whether flocculant needs to be added at the current moment is as follows:

[0026] If the current flocculant addition suitability is greater than or equal to the preset suitability threshold, then no further flocculant needs to be added; otherwise, further flocculant needs to be added.

[0027] Secondly, embodiments of this application also provide a harmless treatment system for enterprise biochemical sludge, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.

[0028] The embodiments of this application have at least the following beneficial effects:

[0029] This application acquires hyperspectral images of the thickener at various times after the addition of flocculant, analyzes the similarity between the hyperspectral data of each pixel and the hyperspectral data of standard flocs to determine the floc pixels in the hyperspectral images at each time point, analyzes the differences in floc pixel distribution between each time point and the first time point to construct the sludge flocculation coefficient at each time point, reflecting the floating status of flocs on the liquid surface in the thickener and evaluating the floc formation on the liquid surface of the thickener, and analyzes in real time the changing trend of the similarity between the hyperspectral images and the hyperspectral images of the liquid surface of the standard thickener with good flocculation effect, as well as the distribution of flocs. Calculating the floc formation effect coefficient at each time point reflects the characteristics of floc suspension changes caused by liquid changes in the thickener. This allows for further assessment of the reaction between sludge and flocs in the thickener, and, combined with the sludge flocculation coefficient, determines whether the flocculant dosage is insufficient and whether further flocculant addition is necessary. After sufficient flocculant is added, the flocs in the thickener are collected through sedimentation. This solves the problem that the flocculant dosage cannot be adjusted in a timely manner according to the actual sludge conditions during the traditional sludge thickening process, thus improving the final sludge flocculation effect and increasing sludge settling treatment efficiency. Attached Figure Description

[0030] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 A flowchart illustrating the steps of a method for harmless treatment of enterprise biochemical sludge, provided in one embodiment of this application;

[0032] Figure 2 This is a schematic diagram illustrating the process of obtaining the sludge flocculation coefficient. Detailed Implementation

[0033] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method and system for the harmless treatment of enterprise biochemical sludge proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0034] Unless otherwise defined, 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 application pertains.

[0035] The following description, in conjunction with the accompanying drawings, details a specific scheme for a method and system for the harmless treatment of enterprise biochemical sludge provided in this application.

[0036] Please see Figure 1 The diagram illustrates a flowchart of a method for harmless treatment of enterprise biochemical sludge according to an embodiment of this application. The method includes the following steps:

[0037] Step S1: Collect hyperspectral image data of the liquid surface in the sludge thickening tank at various times after adding flocculant.

[0038] This application primarily analyzes the single-cycle sludge treatment process of a single gravity thickener. After starting the equipment, a fixed amount of sludge to be treated is added to the first thickener through the sludge inlet. In this application, the addition amount is 300 kg. Implementers should adjust this amount based on the actual single-cycle processing capacity of the first thickener of the gravity thickener; this application does not impose specific limitations. After the sludge addition is complete, the flocculant valve is opened, and a preset amount of flocculant is added to the first thickener through the flocculant injector.

[0039] A hyperspectral camera is installed at the motor position in the middle of the radial truss support bridge on the thickening tank of the gravity thickening equipment. After the flocculant is added, the hyperspectral camera is started and hyperspectral image data of the liquid surface of the first thickening tank is collected vertically downward at intervals T. Standard sludge floc hyperspectral data is obtained from the database connected to the equipment for subsequent processing.

[0040] For the image data acquisition interval, in this embodiment, the value of T is set to 2 seconds. In other embodiments of this application, the implementer may set the value of T according to the actual situation.

[0041] Step S2: Determine the flocculent pixels in the hyperspectral images at each time based on the similarity between the hyperspectral data of each pixel in the hyperspectral images at each time and the standard hyperspectral data; cluster all flocculent pixels in the hyperspectral images at each time, and calculate the sludge flocculation coefficient at the current time based on the number of clusters and the distance between clusters at the current time, combined with the difference between the clustering results at the current time and the first time.

[0042] In the initial stage of sludge settling, gravity thickening equipment adds flocculant to the first thickening tank. The sludge-water mixture in the tank is affected by the flocculant, gradually agglomerating to form flocs and settling. Ultimately, the upper layer of the first thickening tank consists of clear water, and the lower layer consists of sludge flocs. Finally, the sludge flocs settle through the pores at the bottom of the first thickening tank to the second thickening tank for further treatment. However, when the flocculant dosage is insufficient, the flocculation effect decreases, resulting in flocs of varying sizes. Larger flocs gradually settle, while smaller flocs remain suspended or float. Some areas of sludge may fail to form flocs. Ultimately, the upper layer of water in the first thickening tank contains a certain amount of small flocs and some sludge components, with low uniformity of distribution. Specifically, in the hyperspectral image data of the first thickening tank's liquid surface, the overall sludge floc content is high, the distribution is scattered, and the floc size does not increase further over time, eventually settling.

[0043] To characterize the aforementioned features, for each pixel in the hyperspectral image data of the first thickening tank liquid surface at a single moment, combined with the standard hyperspectral data of sludge flocs, if the cosine similarity between the hyperspectral data contained in a single pixel and the standard hyperspectral data of sludge flocs is greater than or equal to a preset similarity threshold, the corresponding position in that pixel is considered to contain sludge flocs and is labeled as a floc pixel; otherwise, the pixel is considered to contain only water and is labeled as a water pixel. In this embodiment, the similarity threshold is set to 0.7. Implementers can adjust the similarity threshold according to the actual thickening situation, and this application does not impose specific restrictions. This method can be used to label each pixel in the hyperspectral image data of the first thickening tank liquid surface at a single moment. Cosine similarity is a well-known technique, and the specific process will not be elaborated further.

[0044] It should be noted that this application provides only one similarity algorithm for calculating the similarity between a single pixel hyperspectral data and standard hyperspectral data. There are many existing similarity algorithms, and implementers may also use other similarity algorithms to calculate the similarity between the two hyperspectral data. This application does not impose any specific restrictions.

[0045] After annotation as described above, all flocculent pixels in the hyperspectral image data at a single time point are used as input. The DPC clustering algorithm is applied, with the distance measured as the Euclidean distance between the flocculent pixel coordinates and a cutoff distance of 3. The output consists of multiple clusters, and the pixels within a single cluster represent the pixels contained at the location of a single flocculent. The DPC clustering algorithm is a well-known technique, and its specific process will not be elaborated upon.

[0046] It should be noted that this application provides only one clustering algorithm for clustering flocculent pixels. There are many existing clustering algorithms, and implementers may also use other clustering algorithms to cluster flocculent pixels. This application does not impose any specific restrictions.

[0047] Based on the above analysis, the sludge flocculation coefficient of the thickener at each time point is characterized by the following relationship:

[0048]

[0049] In the formula, Let be the sludge flocculation coefficient at time i. Let be the variance of the Euclidean distance between any two clusters at time i. Let be the total number of clusters at time i. , These are the average total number of pixels contained in all clusters at the i-th and first time points, respectively. The first time point is the initial time of hyperspectral image data acquisition. This is the first difference.

[0050] It should be noted that, in the embodiments of this application, the Euclidean distance between clusters is the Euclidean distance between the cluster centers; for the measurement of the distance between cluster centers, this application only provides one distance measurement algorithm. There are many existing distance measurement algorithms, and implementers may also use other distance measurement algorithms to calculate the distance between cluster centers. This application does not impose any specific restrictions.

[0051] The meaning of this relationship is: at the i-th time, in the hyperspectral image data of the liquid surface of the first thickening tank, the larger the variance of the Euclidean distance between each cluster, the more clusters there are, and the smaller the change in the size of the clusters over time, it indicates that the flocculation effect of the mud-water mixture in the first thickening tank is poor under the action of flocculant, which may be caused by insufficient flocculant dosage, and the dosage of flocculant needs to be adjusted.

[0052] Step S3: Based on the trend of similarity between the hyperspectral image data of all previous times and the preset standard hyperspectral image data, and combined with the degree of disorder in the distribution of the number of clusters of all previous times, calculate the floc formation effect coefficient of the current time.

[0053] In the actual operation of the gravity thickening equipment, the flocculant in the first thickening tank is mainly added through a flocculant injector that surrounds the tank. After addition, the flocculant gradually mixes with the sludge, thus gradually forming flocs. The number of flocs gradually increases, and when the volume of the flocs increases to a certain extent, they begin to settle, and the number of flocs on the liquid surface gradually decreases. However, since the flocculant injector is located at the edge of the first thickening tank, there is a certain time interval between addition and complete reaction with the mud-water mixture to form flocs and settle. Furthermore, due to the certain deviation in the sludge concentration and composition of the mud-water mixture, the time required for each reaction to complete also varies. Therefore, if the judgment is made solely through step S2, the incomplete reaction may be judged as insufficient flocculant addition, thus controlling the injector to continue adding, ultimately causing problems with the flocculant addition amount. Therefore, further analysis is needed in conjunction with the reaction between the flocculant and the mud-water mixture.

[0054] Specifically, under normal circumstances, after adding sufficient flocculant, the sludge-water mixture will react quickly. The mixture in the upper layer of the thickener will quickly lighten in color due to the formation of flocs from the reaction of sludge and flocculant, eventually becoming clear water. However, since the flocculant is added through the projectors around the thickener, the reaction process will spread from the periphery to the center. When the amount of flocculant added is insufficient, the formation of flocs is slower, the floc volume is smaller, and the mixture in the thickener will have some movement due to mechanical agitation. As a result, the settling speed of the flocs is slower or they may remain suspended on the surface. They may also float back to the surface due to agitation. Consequently, the total number of flocs fluctuates on the surface, and the overall color change of the mixture in the thickener is smaller and slower.

[0055] To characterize the above features, for hyperspectral image data at a single moment, the hyperspectral data of each pixel, i.e., the reflectance of each band contained in each pixel, is used. The fusion result of the same band data of all pixels in the hyperspectral image is used as the value of the corresponding band of the hyperspectral image. The values ​​of all bands of the hyperspectral image are obtained and arranged into a sequence in descending order of bands. This sequence is used as the hyperspectral data sequence of the hyperspectral image data. In this embodiment, the fusion method of the same band data of all pixels is to average the same band data of all pixels. Further, standard hyperspectral image data of the liquid surface in the thickener tank with good floc sedimentation effect when the flocculant dosage is sufficient is obtained, and the above-mentioned same band data fusion of pixels is performed. Finally, the standard thickener tank liquid surface hyperspectral data sequence can be obtained through the above method. The cosine similarity between the hyperspectral data sequence at a single moment and the standard thickener tank liquid surface hyperspectral data sequence is calculated as the reactivity of the hyperspectral image data at the current moment. The larger the cosine similarity between the two, the more appropriate the current flocculant dosage and the better the reactivity. The reactivity of the hyperspectral image data at each time point can be calculated using the above method. Finally, the reactivity of the hyperspectral images at all times is arranged in chronological order to construct a reactivity sequence of the hyperspectral image data.

[0056] Based on the above analysis, the reaction effect of the flocculant in the thickener at the current moment can be characterized by the following relationship:

[0057]

[0058] In the formula, Let be the floc formation effect coefficient at time i; where i represents the order corresponding to time i. , These represent the reactivity of the hyperspectral image data at time j and (j-1) time, respectively. The trend strength of the reactivity sequence of the hyperspectral image data up to the i-th time point; Let the variance of the total number of clusters of flocculent pixels in all hyperspectral image data before the i-th time be denoted as the first variance. This is the second difference.

[0059] It should be noted that the trend intensity in this embodiment is calculated using the trend intensity calculation formula in the STL decomposition algorithm. There are many existing algorithms for calculating trend intensity, and implementers may also use other trend algorithms to obtain trend intensity. This application does not impose any specific restrictions.

[0060] The meaning of this relationship is as follows: When up to the i-th time, the greater the difference between the hyperspectral image data collected at each time and the previous time and the hyperspectral image data of the standard thickener liquid surface, the stronger the trend of the reactivity sequence, that is, the faster the flocculant changes with the sludge in the thickener, and the smaller the variance of the total number of floc clusters, the better the floc formation effect in the current thickener, the faster the floc settling speed, and the faster the water on the surface of the thickener becomes clear water, that is, the current floc dosage is more appropriate.

[0061] Step S4: Determine the appropriateness of flocculant addition at the current moment based on the sludge flocculation coefficient and the floc formation effect coefficient, so as to determine whether flocculant needs to be added at the current moment; after sufficient flocculant is added, the flocs in the thickening tank are settled and collected.

[0062] Based on the above analysis, the appropriateness of flocculant dosage at each time point is calculated using the following expression:

[0063]

[0064] In the formula, To determine the appropriateness of flocculant dosage at time i, Let be the sludge flocculation coefficient at time i. Let floc formation effect coefficient be denoted as i-th time step. This is the normalization function. This is the first product.

[0065] It should be noted that, regarding the normalization method of the first product, this embodiment of the application normalizes the first product at all times by using the maximum-minimum normalization method to obtain the normalized value of the first product at each time. Implementers may also use other normalization methods to obtain the normalized value of the first product, and this application does not impose specific restrictions.

[0066] The meaning of this relationship is: at time i, the higher the degree of sludge flocculation in the thickening tank and the better the floc formation effect, the more appropriate the current dosage of flocculant is and the better the thickening effect. Conversely, it is considered that the current dosage of flocculant is too low, the sludge thickening effect is poor, and there is still a lot of sludge in the water, so further flocculant needs to be added.

[0067] The above method allows for the calculation of flocculant dosing suitability at all times. Hyperspectral image data of the sludge-water mixture surface in the thickening tank from m historical sludge treatments are obtained. This application uses historical data from the past 30 times; however, the implementer can adjust the value of m according to the actual equipment operation, and this application does not impose specific limitations. The above method is used to calculate the flocculant dosing suitability at each time point during each treatment. Finally, all flocculant dosing suitability values ​​are used as input, and cross-validation is employed to output the flocculant dosing suitability threshold for the current time point. If the flocculant dosing suitability at the current time point is greater than or equal to this threshold, it indicates that the current flocculant dosage is appropriate, and subsequent treatment can continue. Conversely, if it is less than this threshold, it indicates that the current flocculant dosage is insufficient, and the flocculant pipeline valve needs to be opened to add more flocculant to the thickening tank. Cross-validation is a known technique, and its specific process will not be elaborated further.

[0068] After sufficient flocculant is added, the flocs in the thickening tank gradually settle into the second thickening tank, while the first thickening tank is replenished with sludge. In the second thickening tank, a scraper is activated to remove the sludge flocs deposited at the bottom. As the flocs in the thickening tank further settle into the second thickening tank, the amount of sludge flocs in the second thickening tank gradually increases. The height of the sludge flocs in the second thickening tank is measured in real time using an ultrasonic level gauge in the radial truss support bridge. When the ultrasonic level gauge indicates that the height of the sludge flocs in the second thickening tank exceeds a preset threshold, the gravity thickening equipment activates the sludge pump at the bottom to extract the sludge from the second thickening tank.

[0069] A schematic diagram of the process for obtaining the sludge flocculation coefficient is shown below. Figure 2 As shown.

[0070] Based on the same inventive concept as the above methods, this application also provides a harmless treatment system for enterprise biochemical sludge, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described methods for harmless treatment of enterprise biochemical sludge.

[0071] In summary, this application provides a method for the harmless treatment of enterprise biochemical sludge. By collecting hyperspectral images of the thickener at various times after the addition of flocculant, the similarity between the hyperspectral data of each pixel and the hyperspectral data of standard flocs is analyzed to determine the floc pixels in the hyperspectral images at each time point. The differences in floc pixel distribution between each time point and the first time point are analyzed to construct the sludge flocculation coefficient at each time point, reflecting the floating status of flocs on the liquid surface in the thickener and evaluating the floc formation on the thickener surface. The method also analyzes in real-time the changing trends in similarity between the hyperspectral images and the hyperspectral images of the standard thickener surface with good flocculation effect. The system analyzes the potential and distribution of flocs, calculates the floc formation effect coefficient at each moment, and reflects the characteristics of floc suspension changes caused by liquid changes in the thickener. This allows for further assessment of the reaction between sludge and flocs in the thickener. Combined with the sludge flocculation coefficient, it determines whether the flocculant dosage is insufficient and whether further flocculant addition is necessary. After sufficient flocculant is added, the flocs in the thickener are collected through sedimentation. This method solves the problem that the flocculant dosage cannot be adjusted in a timely manner according to the actual sludge conditions during the traditional sludge thickening process, thus improving the final sludge flocculation effect and increasing sludge settling treatment efficiency.

[0072] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0073] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0074] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for the harmless treatment of enterprise biochemical sludge, characterized in that, The method includes the following steps: Hyperspectral image data of the liquid level in the sludge thickening tank at various times after the addition of flocculant were collected; Based on the similarity between the hyperspectral data of each pixel in the hyperspectral image at each time and the standard hyperspectral data, the flocculent pixels in the hyperspectral image at each time are determined; all flocculent pixels in the hyperspectral image at each time are clustered, and the sludge flocculation coefficient at the current time is calculated based on the number of clusters at the current time and the distance between clusters, combined with the difference between the clustering results at the current time and the first time. Based on the trend of similarity between the hyperspectral image data of all previous times and the preset standard hyperspectral image data, and combined with the degree of disorder in the distribution of the number of clusters of all previous times, the floc formation effect coefficient of the current time is calculated. The appropriateness of flocculant addition at the current moment is determined based on the sludge flocculation coefficient and the floc formation effect coefficient, so as to determine whether flocculant needs to be added at the current moment; after sufficient flocculant is added, the flocs in the thickening tank are settled and collected. The process for obtaining the sludge flocculation coefficient is as follows: Calculate the variance of the distance between any two clusters at the current time; calculate the mean of the number of pixels in all clusters at each time; calculate the difference between the mean at the current time and the mean at the first time, and denote it as the first difference; The sludge flocculation coefficient at the current moment is positively correlated with the number of clusters and the variance at the current moment, and negatively correlated with the first difference. The process for obtaining the floc formation effect coefficient is as follows: The hyperspectral data sequence of each hyperspectral image data is determined by the hyperspectral data of all pixels in each hyperspectral image data; the similarity of the hyperspectral data sequence between the hyperspectral image data at the current time and the preset standard hyperspectral image data is calculated as the reactivity at the current time; the difference between the reactivity at two adjacent times is calculated and recorded as the second difference. The trend strength is obtained by using the sequence of responsiveness from all previous moments as input to the trend algorithm. Calculate the variance of the number of clusters at all times before the current time, and denote it as the first variance; The floc formation effect coefficient at the current moment is positively correlated with the second difference and the trend intensity, and negatively correlated with the first variance.

2. The method for harmless treatment of enterprise biochemical sludge as described in claim 1, characterized in that, The process of acquiring the flocculent pixels is as follows: Pixels whose similarity to the hyperspectral data is greater than or equal to the preset similarity threshold are designated as floc pixels.

3. The method for harmless treatment of enterprise biochemical sludge as described in claim 1, characterized in that, The distance between clusters is the distance between the center points of the clusters.

4. The method for harmless treatment of enterprise biochemical sludge as described in claim 1, characterized in that, The sludge flocculation coefficient at the current moment is calculated as the product of the number of clusters at the current moment and the variance, multiplied by the first difference.

5. The method for harmless treatment of enterprise biochemical sludge as described in claim 1, characterized in that, The process of acquiring the hyperspectral data sequence is as follows: The fusion result of the same band data of all pixels in the hyperspectral image is used as the value of the corresponding band of the hyperspectral image. The values ​​of all bands of the hyperspectral image are obtained to construct the hyperspectral data sequence of the hyperspectral image.

6. The method for harmless treatment of enterprise biochemical sludge as described in claim 1, characterized in that, The appropriateness of flocculant addition is defined as the normalized value of the product of the sludge flocculation coefficient and the floc formation effect coefficient.

7. The method for harmless treatment of enterprise biochemical sludge as described in claim 1, characterized in that, The process of determining whether flocculant needs to be added at the current moment is as follows: If the current flocculant addition suitability is greater than or equal to the preset suitability threshold, then no further flocculant needs to be added; otherwise, further flocculant needs to be added.

8. A system for the harmless treatment of enterprise biochemical sludge, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-7.

Images