Analysis Method, Device, Equipment and Storage Medium for Industrial Solid Waste Data

Abstract

The present invention relates to the technical field of data analysis and processing, in particular to an analysis method, device, equipment and storage medium for industrial solid waste data. The method monitors data in real time by setting a sliding detection window on the data stream, initially selects abnormal data and increases the length of the sliding detection window. In this way, at the preset end time, the abnormal industrial solid waste data can be further confirmed by the difference between the length of the sliding detection window and the data size in the data buffer. Compared with the prior art, the present invention uses the growth rate of the sliding detection window and the data increase rate in the data buffer as further verification after initially screening abnormal data, effectively and quickly selects abnormal industrial solid waste data from the data stream, is particularly suitable for occasions with high real-time requirements, and solves the problem of effectively removing abnormal data in industrial solid waste data while ensuring real-time performance in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis and processing, and particularly to an analysis method, device, equipment and storage medium for industrial solid waste data. Background Art

[0002] Industrial solid waste refers to the waste generated during the production process, which has a wide variety of types, including wastewater, waste gas, waste residue, etc. The formation of industrial solid waste mainly stems from the waste and by-products of production activities and occupies an important position in industrial production. The reasonable treatment and management of industrial solid waste is an important task in the industrial production process, which is directly related to the protection of the production environment and the sustainable utilization of resources.

[0003] Industrial solid waste data refers to the data collected during the industrial production process, including information such as the types of solid waste, the generation amount, and the treatment methods. Systematic analysis of industrial solid waste data can help enterprises understand the situation of solid waste generated during their production process, so as to optimize the production process targeted, improve the resource utilization efficiency, and reduce the impact on the environment. At the same time, analyzing industrial solid waste data can also provide a scientific basis for enterprises to formulate reasonable solid waste treatment plans and promote the transformation of industrial production towards sustainable development.

[0004] Data quality is a crucial issue in the analysis of industrial solid waste data, and its effect directly affects the accuracy and reliability of subsequent analysis. However, the current industrial solid waste data is large in scale and complex in information, which brings great challenges to data cleaning. The current data cleaning methods include clustering-based methods, data normalization, outlier detection, etc. Although they can ensure the accuracy of data cleaning, the cleaning speed is slow and it is difficult to meet the application scenarios with high real-time requirements, such as monitoring the generation situation of solid waste during the production process and adjusting the treatment plan in real time. Therefore, how to effectively remove the abnormal data in industrial solid waste data while ensuring real-time performance is an urgent problem that people hope to solve. Summary of the Invention

[0005] Therefore, the present invention provides an analysis method, device, equipment and storage medium for industrial solid waste data to solve the problem of how to effectively remove the abnormal data in industrial solid waste data while ensuring real-time performance in the prior art.

[0006] The present invention provides an analysis method for industrial solid waste data, including:

[0007] Obtain industrial solid waste data in real time, and at a preset start time, set a sliding detection window with a preset length on the data stream composed of industrial solid waste data;

[0008] Based on a preset abnormal data standard, abnormal data is selected from the industrial solid waste data covered by the sliding detection window, and the abnormal data is saved to the data buffer;

[0009] The length of the sliding detection window is increased according to the data size in the data buffer, wherein the increasing speed of the length of the sliding detection window is less than the increasing speed of the data in the data buffer;

[0010] The sliding detection window is moved backward, and the step of selecting abnormal data from the industrial solid waste data covered by the sliding detection window based on the preset abnormal data standard and saving the abnormal data to the data buffer is returned for execution;

[0011] At the preset end time, if the difference between the length of the sliding detection window and the data size in the data buffer is greater than the preset threshold, the data in the data buffer is used as abnormal industrial solid waste data.

[0012] Further, the length of the sliding detection window is increased based on the following formula:

[0013] l = l 0 + 1 ln( 2 s + 1);

[0014] wherein, l represents the increased length of the sliding detection window, l 0 represents the preset length, ln() is the natural logarithm function, s is the data size in the data buffer, ω 1 is the proportional adjustment coefficient, ω 2 is the unit adjustment coefficient.

[0015] Further, it further includes:

[0016] At the preset end time, if the difference between the length of the sliding detection window and the data size in the data buffer is less than or equal to the preset threshold, the data buffer is cleared and the length of the sliding detection window is reset to the preset length.

[0017] Further, the preset abnormal data standard includes an abnormal data standard value, and at the preset start time, the abnormal data standard value is the initial value;

[0018] Before performing the step of moving the sliding detection window backward, the analysis method of the industrial solid waste data further includes:

[0019] According to the statistical characteristics of the industrial solid waste data covered by the sliding detection window, an updated value is calculated;

[0020] The abnormal data standard value is set to the updated value;

[0021] After the preset end time, the analysis method of the industrial solid waste data further includes:

[0022] If the difference between the length of the sliding detection window and the size of the data in the data buffer is less than or equal to a preset threshold, reset the abnormal data standard value to the initial value.

[0023] Furthermore, the industrial solid waste data includes waste emission concentration and waste emission volume, and the abnormal data standard value includes a preset total threshold representing the product of the waste emission concentration and the waste emission volume; the selecting abnormal data from the industrial solid waste data covered by the sliding detection window based on the preset abnormal data standard includes:

[0024] Obtain the industrial solid waste data covered by the sliding detection window to obtain a time-waste emission concentration curve and a time-waste emission volume curve;

[0025] According to the product relationship between the time-waste emission concentration curve and the time-waste emission volume curve, obtain a time-total curve;

[0026] Compare the time-total curve with the preset total threshold, and take the points on the time-total curve where the total exceeds the preset total threshold as abnormal data.

[0027] Furthermore, the calculating the update value according to the statistical characteristics of the industrial solid waste data covered by the sliding detection window includes:

[0028] Obtain the time stamp corresponding to the last abnormal data stored in the data buffer;

[0029] According to the time stamp corresponding to the abnormal data, calculate the change rate of the waste emission concentration and the change rate of the waste emission volume at the moment corresponding to this time stamp based on the time-waste emission concentration curve and the time-waste emission volume curve;

[0030] Calculate the update value according to the change rate of the waste emission concentration and the change rate of the waste emission volume.

[0031] Furthermore, the calculation formula of the update value is as follows:

[0032] g = g 0 -(a(ΔC) 2 +bΔCΔE+c() 2 );

[0033] where, g represents the update value, g 0 represents the initial value, |ΔC| represents the absolute value of the change rate of the waste emission concentration, |ΔC| represents the absolute value of the change rate of the waste emission volume, and a, b, and c respectively represent different adjustment weight parameters.

[0034] The present invention also provides an analysis device for industrial solid waste data, including:

[0035] A window setting module, configured to obtain industrial solid waste data in real time, and set a sliding detection window with a preset length on the data stream formed by the industrial solid waste data at a preset start time;

[0036] A primary detection module, configured to select abnormal data from the industrial solid waste data covered by the sliding detection window based on a preset abnormal data standard, and save the abnormal data to a data buffer;

[0037] A window adjustment module, configured to increase the length of the sliding detection window according to the data size in the data buffer, wherein the increasing speed of the length of the sliding detection window is less than the increasing speed of the data in the data buffer;

[0038] A window moving module, configured to move the sliding detection window backward, and return to execute the step of selecting abnormal data from the industrial solid waste data covered by the sliding detection window based on a preset abnormal data standard, and saving the abnormal data to the data buffer;

[0039] A secondary detection module, configured to, at a preset end time, if the difference between the length of the sliding detection window and the data size in the data buffer is greater than a preset threshold, use the data in the data buffer as abnormal industrial solid waste data.

[0040] The present invention also provides an analysis device for industrial solid waste data, including:

[0041] A memory and a processor;

[0042] The memory is used to store a program, and the processor is configured to execute the steps in the analysis method of the industrial solid waste data described in any one of the above when executing the program.

[0043] The present invention also provides a storage medium, configured to store a computer-readable program or instruction, and when the program or instruction is executed by a processor, it can implement the steps in the analysis method of the industrial solid waste data described in any one of the above.

[0044] The beneficial effects of adopting the above embodiments are:

[0045] The present invention monitors data in real time by setting a sliding detection window on the data stream. At a preset start time, abnormal data is initially selected according to a preset abnormal data standard and saved in a data buffer. Then, as the window moves, the above actions are repeated. During the process of initially selecting abnormal data, the length of the sliding detection window is increased according to the data size in the data buffer. In this way, at the preset end time, since the growth rate of the length of the sliding detection window is less than the growth rate of the data in the data buffer, when the growth rate of the sliding detection window is relatively close to the growth rate of the data in the data buffer, it can be considered that all the newly detected data by the sliding detection window has been put into the data buffer. At this time, although the industrial solid waste data is different from the preset abnormal standard, it is a normal fluctuation of the data itself rather than abnormal data. On the contrary, if within a period of time until the preset end time, the difference between the length of the sliding detection window and the data size in the data buffer is greater than a preset threshold, it can be considered that the growth rate of the data in the data buffer has not caught up with the growth rate of the sliding detection window. Then, at this time, it can be considered that the previously initially selected abnormal data is indeed abnormal industrial solid waste data with a high degree of difference from other data. Compared with the prior art, the present invention uses the growth rate of the sliding detection window and the growth rate of the data in the data buffer as a further verification after initially screening abnormal data, without the need for a complex comparison operation process, without storing a large amount of data for further processing, effectively and quickly selecting abnormal industrial solid waste data from the data stream, and is particularly suitable for occasions with high real-time requirements, solving the problem of how to effectively eliminate abnormal data in industrial solid waste data while ensuring real-time performance in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flowchart of a method of an embodiment of the method for analyzing industrial solid waste data provided by the present invention;

[0047] Figure 2 It is a flowchart of a method of another embodiment of the method for analyzing industrial solid waste data provided by the present invention;

[0048] Figure 3 is Figure 1 a flowchart of a method of an embodiment of step S102 in

[0049] Figure 4 It is a structural schematic diagram of an analysis device for industrial solid waste data provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present invention.

[0051] In combination with Figure 1 As shown, a specific embodiment of the present invention discloses a method for analyzing industrial solid waste data, including:

[0052] S101. Obtain industrial solid waste data in real time, and at a preset start time, set a sliding detection window with a preset length on the data stream composed of industrial solid waste data;

[0053] S102. Based on a preset abnormal data standard, select abnormal data from the industrial solid waste data covered by the sliding detection window, and save the abnormal data to a data buffer;

[0054] S103. Increase the length of the sliding detection window according to the data size in the data buffer, wherein the increasing speed of the length of the sliding detection window is less than the increasing speed of the data in the data buffer;

[0055] S104. Move the sliding detection window backward, and return to execute the step of selecting abnormal data from the industrial solid waste data covered by the sliding detection window based on a preset abnormal data standard, and saving the abnormal data to the data buffer;

[0056] S105. At a preset end time, if the difference between the length of the sliding detection window and the data size in the data buffer is greater than a preset threshold, then use the data in the data buffer as abnormal industrial solid waste data.

[0057] It should be noted that the abnormal data and abnormal industrial solid waste data herein both refer to "error" data, that is, the data changes are random irregular change data generated by factors such as signal jumps and sensor jitters. These data are different from the abnormal data caused by factors such as process changes and operation errors. Although the abnormal data caused by factors such as process changes and operation errors do not meet the relevant indicators, these data can still reflect the influence relationship between different factors and still have analysis significance, and should be retained in the subsequent analysis work. And the abnormal data herein is error data that has no association with other data and does not carry meaningful information itself. Therefore, it needs to be promptly removed during data cleaning to avoid damaging the subsequent analysis work. It can be understood that the focus of the present invention is on how to determine abnormal industrial solid waste data, and how to perform the next processing on these data is not within the scope of discussion of the present invention.

[0058] In addition, the sliding detection window in the above process is a sliding window for detection. The sliding window is an algorithm technique used to perform operations on continuous subsequences on a sequence or an array. Its basic idea is to maintain a window of a fixed size and slide the window on the sequence to perform a certain operation, such as calculating the sum of a subsequence, finding the maximum or minimum value, etc. It can be understood that the sliding detection window, the data buffer, and the related operation adjectives in this embodiment are prior arts whose meanings can be clearly understood by those skilled in the art, so no further description will be given herein.

[0059] The preset start time and the preset end time in the above process need to be set according to specific circumstances. The two can be the start time and the end time of a day, or the start time and the end time of a shorter time such as an hour.

[0060] The present invention performs real-time monitoring on data by setting a sliding detection window on the data stream. At the preset start time, abnormal data is initially selected according to the preset abnormal data standard and saved to the data buffer, and then the above actions are repeated as the window moves. During the process of initially selecting abnormal data, the length of the sliding detection window is increased according to the data size in the data buffer. In this way, at the preset end time, because the increase speed of the length of the sliding detection window is less than the increase speed of the data in the data buffer, when the growth speed of the sliding detection window is relatively close to the increase speed of the data in the data buffer, it can be considered that all the newly detected data by the sliding detection window has been put into the data buffer. At this time, although the industrial solid waste data is different from the preset abnormal standard, it is the normal fluctuation of the data itself rather than abnormal data; on the contrary, if within a period of time until the preset end time, the difference between the length of the sliding detection window and the data size in the data buffer is greater than the preset threshold, it can be considered that the increase speed of the data in the data buffer has not caught up with the growth speed of the sliding detection window. At this time, it can be considered that the abnormal data initially selected before is indeed abnormal industrial solid waste data with a high degree of difference from other data.

[0061] Compared with the prior art, the present invention uses the growth speed of the sliding detection window and the increase speed of the data in the data buffer as a further verification after initially screening abnormal data, without the need for complex comparison operation processes, without storing a large amount of data for further processing, effectively and quickly selecting abnormal industrial solid waste data from the data stream, and is particularly suitable for occasions with high real-time requirements, solving the problem of how to effectively eliminate abnormal data in industrial solid waste data while ensuring real-time performance in the prior art.

[0062] In the above step S103, it is necessary to adjust the sliding detection window. The expansion of the sliding detection window starts when the first abnormal data is obtained. In addition to comparing with the data size in the data buffer, the most direct effect of expanding the sliding detection window is to expand the detection range of industrial solid waste data, enabling it to grasp the overall law of data within a wider field of vision, thereby improving the accuracy of finding abnormal data. At the same time, expanding the sliding detection window also provides a technical basis for adjusting the preset abnormal data standard subsequently, enabling the preset abnormal data standard to be adjusted dynamically during the operation of the method, making it more in line with the current actual situation and further improving the accuracy.

[0063] Furthermore, in a preferred embodiment, the length of the sliding detection window is increased based on the following formula:

[0064] l = l 0 + 1 ln( 2 s + 1);

[0065] where, l represents the length of the sliding detection window after increase, l 0 represents the preset length, ln() is the natural logarithm function, s is the data size in the data buffer, ω 1 is the proportional adjustment coefficient, and ω 2 is the unit adjustment coefficient.

[0066] The significance of the above formula is to determine a suitable growth rate according to the data size in the data buffer, so as to obtain the length of the sliding detection window after increase. In addition, it should be noted that this formula is designed using the natural logarithm function, making the sliding detection window sensitive to the appearance of abnormal data. That is, when the data in the data buffer is less, the sliding detection window can grow at a faster speed to quickly expand the detection field of vision and analyze the overall data law to ensure accurate identification. When the data in the data buffer is more, at this time the sliding detection window has reached a relatively large level, and continuing to increase the length of the sliding detection window will increase the operating pressure of the computer and reduce the real-time performance of the entire algorithm. Therefore, at this time, the increase of the sliding detection window should be appropriately slowed down to ensure the data processing speed.

[0067] Furthermore, as shown in Figure 2 , in a preferred embodiment, the analysis method of the industrial solid waste data further includes:

[0068] S106. At the preset end time, if the difference between the length of the sliding detection window and the data size in the data buffer is less than or equal to the preset threshold, clear the data buffer and reset the length of the sliding detection window to the preset length.

[0069] It can be understood that the above is only a preferred implementation manner. In practice, the data in the data buffer can also be cleared one by one at regular time intervals.

[0070] Further, please refer to Figure 2 again. The preset abnormal data standard includes an abnormal data standard value, which is the initial value at the preset start time.

[0071] Before performing the step of moving the sliding detection window backward, the analysis method of the industrial solid waste data further includes:

[0072] S107. Calculate an update value according to the statistical characteristics of the industrial solid waste data covered by the sliding detection window;

[0073] S108. Set the abnormal data standard value to the update value;

[0074] After the preset end time, the analysis method of the industrial solid waste data further includes:

[0075] S109. If the difference between the length of the sliding detection window and the size of the data in the data buffer is less than or equal to the preset threshold, reset the abnormal data standard value to the initial value.

[0076] In the above process, the abnormal data standard value is a numerical value for comparison. For example, data exceeding the abnormal data standard value is abnormal data. In practice, according to the types of industrial solid waste data and different specific requirements, the preset abnormal data standard can include multiple abnormal data standard values.

[0077] This embodiment further realizes the dynamic adjustment of the preset abnormal data standard, which is especially suitable for some occasions without fixed discrimination indicators, enabling the preset abnormal data standard to change with the real-time changes of industrial solid waste data, so as to achieve a more practical effect.

[0078] Specifically, in a preferred embodiment, the industrial solid waste data includes waste emission concentration and waste emission volume, and the abnormal data standard value includes a preset total threshold representing the product of the waste emission concentration and the waste emission volume. That is, the situation of harmful substances discharged can be measured by the waste emission concentration and the waste emission volume. In this scenario, as shown in Figure 3 , the above step S102, selecting abnormal data from the industrial solid waste data covered by the sliding detection window based on the preset abnormal data standard, specifically includes:

[0079] S301. Obtain the industrial solid waste data covered by the sliding detection window to obtain a time-waste emission concentration curve and a time-waste emission volume curve;

[0080] S302. Obtain the time-total amount curve according to the product relationship between the time-waste emission concentration curve and the time-waste emission amount curve;

[0081] S303. Compare the time-total amount curve with a preset total amount threshold, and take the points on the time-total amount curve where the total amount exceeds the preset total amount threshold as abnormal data.

[0082] Suppose at a certain moment, there is a wrong jump in the waste emission concentration with a sudden increase in value, and at the same time, there is a wrong jump in the waste emission amount with a sudden decrease in value. However, at this time, the product of the two is still below the preset total amount threshold. Obviously, this abnormal data cannot be identified at this time. Then it is obviously inappropriate to continue to maintain the preset abnormal data standard unchanged at this time, and the abnormal data standard value should be appropriately reduced to make the detection standard more stringent.

[0083] For example, in a preferred embodiment, the above step S107, calculating the update value according to the statistical characteristics of the industrial solid waste data covered by the sliding detection window, specifically includes:

[0084] Obtain the timestamp corresponding to the last abnormal data stored in the data buffer;

[0085] According to the timestamp corresponding to the abnormal data, calculate the change rate of the waste emission concentration and the change rate of the waste emission amount at the moment corresponding to this timestamp based on the time-waste emission concentration curve and the time-waste emission amount curve;

[0086] Calculate the update value according to the change rate of the waste emission concentration and the change rate of the waste emission amount.

[0087] The above process incorporates the consideration of the change rate, and controls the mutual relationship between the two from two scales through the change rate of the waste emission concentration and the change rate of the waste emission amount, so as to achieve the effect of adjusting the preset abnormal data standard according to the fluctuation conditions of the time-waste emission concentration curve and the time-waste emission amount curve.

[0088] Specifically, in a preferred embodiment, the calculation formula of the update value is as follows:

[0089] g = g 0 -(a(ΔC) 2 +bΔCΔE+c() 2 );

[0090] Where, g represents the update value, g 0 represents the initial value, |ΔC| represents the absolute value of the change rate of the waste emission concentration, |ΔE| represents the absolute value of the change rate of the waste emission amount, and a, b, and c respectively represent different adjustment weight parameters.

[0091] The significance of the above formula is that when one of the time-waste emission concentration curve and the time-waste emission amount curve fluctuates, the impact on the updated value is less, that is, the preset abnormal data standard will be changed slightly, and when the time-waste emission concentration curve and the time-waste emission amount curve fluctuate simultaneously, the impact on the updated value is greater, that is, the preset abnormal data standard will be changed significantly, thereby improving the accuracy.

[0092] Combined with Figure 4 As shown, the present invention also provides an analysis device 400 for industrial solid waste data, including:

[0093] A window setting module 410, configured to obtain industrial solid waste data in real time, and set a sliding detection window with a preset length on the data stream formed by the industrial solid waste data at a preset start time;

[0094] A primary detection module 420, configured to select abnormal data from the industrial solid waste data covered by the sliding detection window based on a preset abnormal data standard, and save the abnormal data to a data buffer;

[0095] A window adjustment module 430, configured to increase the length of the sliding detection window according to the data size in the data buffer, wherein the increasing speed of the length of the sliding detection window is less than the increasing speed of the data in the data buffer;

[0096] A window moving module 440, configured to move the sliding detection window backward, and return to execute the step of selecting abnormal data from the industrial solid waste data covered by the sliding detection window based on a preset abnormal data standard, and saving the abnormal data to the data buffer;

[0097] A secondary detection module 450, configured to, at a preset end time, if the difference between the length of the sliding detection window and the data size in the data buffer is greater than a preset threshold, use the data in the data buffer as abnormal industrial solid waste data.

[0098] It should be noted here that: the corresponding analysis device 400 for industrial solid waste data provided in the above embodiments can implement the technical solutions described in the above method embodiments. The specific implementation principles of the above modules or units can be referred to the corresponding content in the above method embodiments, and will not be elaborated here.

[0099] The present invention also provides an analysis device for industrial solid waste data. In this embodiment, the electronic device includes:

[0100] A memory and a processor;

[0101] The memory is used to store a program, and the processor is used to execute the analysis method of industrial solid waste data in the above embodiments when executing the program.

[0102] This embodiment also provides a storage medium, on which an analysis program for industrial solid waste data is stored. When the analysis program for industrial solid waste data is executed by a processor, the steps in the above embodiment can be implemented.

[0103] In the present invention, by setting a sliding detection window on the data stream, the data is monitored in real time. At a preset start time, abnormal data is initially selected according to a preset abnormal data standard and saved to the data buffer area. Then, as the window moves, the above actions are repeated. During the process of initially selecting abnormal data, the length of the sliding detection window is increased according to the data size in the data buffer area. In this way, at a preset end time, because the increasing speed of the length of the sliding detection window is less than the increasing speed of the data in the data buffer area, when the increasing speed of the sliding detection window is relatively close to the increasing speed of the data in the data buffer area, it can be considered that all the newly detected data by the sliding detection window has been put into the data buffer area. At this time, although the industrial solid waste data is different from the preset abnormal standard, it is the normal fluctuation of the data itself rather than abnormal data. On the contrary, if within a period of time until the preset end time, the difference between the length of the sliding detection window and the data size in the data buffer area is greater than a preset threshold, it can be considered that the increasing speed of the data in the data buffer area has not caught up with the increasing speed of the sliding detection window. At this time, it can be considered that the abnormal data initially selected before is indeed abnormal industrial solid waste data with a high degree of difference from other data. Compared with the prior art, the present invention uses the increasing speed of the sliding detection window and the increasing speed of the data in the data buffer area as the further verification after initially screening abnormal data, without the need for a complex comparison operation process, without storing a large amount of data for further processing, effectively and quickly selecting abnormal industrial solid waste data from the data stream, especially suitable for occasions with high real-time requirements, and solving the problem of how to effectively eliminate abnormal data in industrial solid waste data while ensuring real-time performance in the prior art.

[0104] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other.

[0105] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for analyzing industrial solid waste data, characterized in that: include: Acquire industrial solid waste data in real time, and set a sliding detection window of a preset length on a data stream consisting of the industrial solid waste data at a preset start time; Based on the preset abnormal data standard, the abnormal data is selected from the industrial solid waste data covered by the sliding detection window, and the abnormal data is saved in the data cache area; Increasing the length of the sliding detection window according to the size of the data in the data buffer area, wherein the speed of increasing the length of the sliding detection window is lower than the speed of increasing the data in the data buffer area; Move the sliding detection window backward, and return to execute the step of selecting abnormal data from the industrial solid waste data covered by the sliding detection window based on a preset abnormal data standard, and saving the abnormal data to a data buffer area; At the preset end time, if the difference between the length of the sliding detection window and the data size in the data cache area is greater than a preset threshold, the data in the data cache area is regarded as abnormal industrial solid waste data.

2. The method for analyzing industrial solid waste data according to claim 1, characterized in that: The length of the sliding detection window increases based on the following formula: ； in, Indicates the length of the increased sliding detection window, Indicates the preset length, is the natural logarithm function, is the data size in the data buffer, is the proportional adjustment coefficient, is the unit adjustment factor.

3. The method for analyzing industrial solid waste data according to claim 1, characterized in that: Also includes: At the preset end time, if the difference between the length of the sliding detection window and the data size in the data buffer is less than or equal to the preset threshold, the data buffer is cleared and the length of the sliding detection window is reset to the preset length.

4. The method for analyzing industrial solid waste data according to claim 1, characterized in that: The preset abnormal data standard includes an abnormal data standard value, and at the preset start time, the abnormal data standard value is an initial value; Before executing the step of moving the sliding detection window backward, the industrial solid waste data analysis method further includes: Calculate the update value according to the statistical characteristics of the industrial solid waste data covered by the sliding detection window; Set the abnormal data standard value to the updated value; After the preset end time, the industrial solid waste data analysis method further includes: If the difference between the length of the sliding detection window and the size of the data in the data buffer is less than or equal to a preset threshold, the abnormal data standard value is reset to an initial value.

5. The method for analyzing industrial solid waste data according to claim 4, characterized in that: Industrial solid waste data include waste emission concentration and waste emission amount, and the abnormal data standard value includes a preset total amount threshold representing the product of waste emission concentration and waste emission amount; The method of selecting abnormal data from the industrial solid waste data covered by the sliding detection window based on the preset abnormal data standard includes: Obtain the industrial solid waste data covered by the sliding detection window to obtain the time-waste emission concentration curve and the time-waste emission amount curve; According to the product relationship between the time-waste discharge concentration curve and the time-waste discharge amount curve, the time-total amount curve is obtained; The time-total amount curve is compared with the preset total amount threshold, and the point in the time-total amount curve where the total amount exceeds the preset total amount threshold is taken as abnormal data.

6. The method for analyzing industrial solid waste data according to claim 5, characterized in that: The step of calculating the update value according to the statistical characteristics of the industrial solid waste data covered by the sliding detection window includes: Get the timestamp corresponding to the last abnormal data stored in the data buffer; According to the timestamp corresponding to the abnormal data, the change rate of the waste emission concentration and the change rate of the waste emission amount at the moment corresponding to the timestamp are calculated based on the time-waste emission concentration curve and the time-waste emission amount curve; The updated value is calculated based on the rate of change of waste emission concentration and the rate of change of waste emission amount.

7. The method for analyzing industrial solid waste data according to claim 6, characterized in that: The calculation formula of the update value is as follows: ； in, Indicates the updated value. represents the initial value, Indicates the absolute value of the rate of change of waste discharge concentration, Indicates the absolute value of the rate of change of waste discharge, , and They represent different adjustment weight parameters respectively.

8. An analysis device for industrial solid waste data, characterized in that: include: A window setting module, used to obtain industrial solid waste data in real time, and set a sliding detection window of a preset length on a data stream consisting of the industrial solid waste data at a preset start time; A primary detection module, for selecting abnormal data from the industrial solid waste data covered by the sliding detection window based on a preset abnormal data standard, and saving the abnormal data to a data cache area; A window adjustment module, used to increase the length of the sliding detection window according to the size of the data in the data buffer area, wherein the speed of increasing the length of the sliding detection window is lower than the speed of increasing the data in the data buffer area; A window moving module, used for moving the sliding detection window backwards, and returning to execute the step of selecting abnormal data from the industrial solid waste data covered by the sliding detection window based on a preset abnormal data standard, and saving the abnormal data to a data buffer area; The secondary detection module is used to treat the data in the data cache area as abnormal industrial solid waste data if the difference between the length of the sliding detection window and the data size in the data cache area is greater than a preset threshold at the preset end time.

9. An analysis device for industrial solid waste data, characterized in that: include: Memory and processor; The memory is used to store programs, and the processor is used to execute the steps in the industrial solid waste data analysis method described in any one of claims 1 to 7 when executing the program.

10. A storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the industrial solid waste data analysis method described in any one of claims 1 to 7.

Images