Industrial park environment brain platform

Abstract

The invention relates to the technical field of environmental protection, and particularly discloses an industrial park environment brain platform, and the platform comprises a data collection end which is arranged at each drainage point of a park and is used for obtaining first sewage data of each drainage point at a fixed interval time point; the data processing end is used for processing the first sewage data to obtain second sewage data; the AI environment brain is used for performing traceability analysis on the first sewage data and the second sewage data to obtain a first traceability result and a second traceability result; the analysis end is used for obtaining a park pollution traceability result according to comparison, judgment and analysis of the first traceability result and the second traceability result; according to the system, the selection of the traceability result is determined according to the comparison, judgment and analysis of the first traceability result and the second traceability result, the data characteristics of the first sewage data and the second sewage data can be combined, the accuracy of the traceability result is ensured, and the comprehensiveness of the judgment result is ensured.

Description

Technical Field

[0001] The present invention relates to the field of environmental protection technology, and in particular to an industrial park environmental brain platform. Background Art

[0002] With the rapid development of AI technology, its application in the field of environmental protection is becoming more and more extensive. Among them, the AI ​​environmental brain is an analysis model based on big data and machine learning. Based on the monitored water quality data and other environmental data, the AI ​​environmental brain can trace back to a specific enterprise while satisfying the overall control of the analog proportion of pollution sources, realize regional pollution one by one, and assess the risk level according to the nature of the pollution, timely discover and curb pollution, and play an important role and value in providing urban environmental managers with accurate tracing, auxiliary decision-making, command and dispatch, and improving intelligent and efficient management.

[0003] In the existing technology, by applying the AI ​​environmental brain to the park environment platform and using the water quality monitoring data uploaded to the platform, the park's environmental status and environmental risks can be promptly judged and the results traced, which greatly facilitates the park's environmental protection department to monitor and manage the wastewater discharge status of enterprises.

[0004] Since the accuracy of the traceability of the AI ​​environmental brain results is mainly based on the accuracy of the water quality monitoring results, during the water quality monitoring process, if the data detected at a single time point deviates from the actual data, the error will be amplified during the judgment process of the AI ​​environmental brain, which will lead to incorrect judgment of the traceability results and interfere with the judgment of managers. At the same time, if the water quality data at multiple time points are processed and input into the AI ​​environmental brain, it will cause some data loss, thereby also reducing the accuracy of the traceability result judgment. Summary of the invention

[0005] The purpose of the present invention is to provide an industrial park environment brain platform to solve the following technical problems:

[0006] How to improve the accuracy of AI environmental brain in judging and tracing environmental risks in industrial parks.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] An industrial park environment brain platform, the platform comprising:

[0009] The data collection terminal is set at each drainage point in the park to obtain the first sewage data at fixed intervals of each drainage point;

[0010] A data processing end, used for processing the first sewage data to obtain second sewage data;

[0011] The AI ​​environment brain is used to perform source tracing analysis on the first sewage data and the second sewage data respectively to obtain a first source tracing result and a second source tracing result;

[0012] The analysis end is used to obtain the pollution source tracing results of the industrial park based on the comparison and analysis of the first source tracing results and the second source tracing results.

[0013] Furthermore, the process of processing the first sewage data includes:

[0014] When the concentration of any detection factor in the first sewage data is monitored to exceed the threshold, the corresponding time point is used as the starting point of the tracing period;

[0015] Obtain the concentration value of each detection factor in the first sewage data during the tracing period and perform fitting to obtain the concentration change curve of each detection factor;

[0016] By formula:

[0017]

[0018] Calculate the concentration value C of the i-th detection factor at the j-th time point i (t j ), the concentration values ​​of all detection factors are taken as the second sewage data;

[0019] Among them, i is the order of detection factor types, j is the order of fixed interval time points, C i (t) is the concentration curve fitted by the i-th detection factor, t j is the jth time point, and Δt is a preset fixed period.

[0020] Furthermore, the first tracing result includes a first venture enterprise set R1, and the second tracing result includes a second venture enterprise set R2;

[0021] The process of comparison, judgment and analysis includes:

[0022] If R1=R2, all enterprises in the first risk enterprise set R1 and the second risk enterprise set R2 are judged as the pollution source tracing results of the park;

[0023] If R1≠R2, the enterprise corresponding to R1∩R2 is judged as the pollution source tracing result of the industrial park, and all enterprises in the set R1-R2 and the set R2-R1 are matched and judged. According to the judgment result, it is determined whether to use the enterprises in the set R1-R2 and the set R2-R1 as the pollution source tracing result of the industrial park.

[0024] Furthermore, the matching judgment process includes:

[0025] Obtain the pollutant sequences of the untreated sewage of enterprises in the set R1-R2 and the set R2-R1 respectively;

[0026] By formula:

[0027]

[0028] Calculate the matching degree G between the x-th enterprise and the current test result x ;

[0029] If the matching degree G x ≥G1, the xth enterprise will be taken as the pollution source tracing result of the park;

[0030] Among them, M x The number of enterprises selected according to the relative emission concentration ranking of the detection factor for the xth enterprise, y∈[1, M x ]; sort xy (t j ) is the yth detection factor of the xth enterprise at t j The relative concentration ranking of time points, f s is a comparison table function, which is a decreasing function, and sort xy (t j )>M x When f s (sort xy (t j ))=0; For all time points f s (sort xy (t j ))’s mean, C yx (t j ) is the yth detection factor of the xth enterprise at t j The value at the time point, C0 y is the reference value of the yth detection factor, Ct y is the standard value of the yth detection factor, and G1 is the matching threshold.

[0031] Furthermore, the process of processing the first sewage data also includes:

[0032] Get the slope k of the concentration curve fitted by the i-th detection factor at each time point ij ;

[0033] The analysis end is also used to calculate the slope k j , concentration value C i (t j ) and execute early warning strategies at corresponding time points.

[0034] Furthermore, the early warning strategy includes:

[0035] By formula:

[0036]

[0037] Calculate and obtain the warning value W(H);

[0038] When the warning value W(H)≥W1, a warning is issued;

[0039] Among them, W1 is the warning threshold, γ is the adjustment coefficient, C0 i is the reference value of the ith detection factor, Ct i is the standard value of the i-th detection factor, N is the number of detection factors, i∈[1,N]; α i is the weight coefficient of the i-th detection factor, H is the number of time points, and H is the variable.

[0040] Furthermore, the early warning strategy also includes:

[0041] When W0≤W(H)<W1, a consistency judgment is made based on the status of each drainage point in the park, and whether to issue an early warning is determined based on the result of the consistency judgment;

[0042] Among them, W0 is the warning risk threshold.

[0043] Furthermore, the consistency judgment process includes:

[0044] By formula:

[0045]

[0046] Calculate the risk consistency coefficient s of all drainage points in the park W ;

[0047] If the risk consistency coefficient s W ≤s1, no warning will be issued;

[0048] If W >s1, an early warning is issued;

[0049] Among them, s1 is the consistency reference threshold, Q is the number of drainage points in the park, z∈[1,Q], W z (H) is the warning value of the zth drainage point at the Hth time point, It is the average of the warning values ​​of all drainage points.

[0050] Beneficial effects of the present invention:

[0051] (1) The present invention determines the selection of the source tracing result based on the comparison and judgment analysis of the first source tracing result and the second source tracing result, thereby obtaining the pollution source tracing result of the industrial park. The present invention can combine the characteristics of the first sewage data and the second sewage data, thereby ensuring the accuracy of the source tracing result and the comprehensiveness of the judgment result.

[0052] (2) The present invention issues an early warning at a more adaptive time point based on the actual situation of the water quality detection risk. Compared with the method of immediately issuing an early warning at the early stage of abnormal water quality parameters, this method can greatly reduce the problem of increased workload of management personnel due to data misjudgment, while ensuring the timeliness of the early warning, thereby making it easier to deal with environmental problems that arise. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The present invention will be further described below in conjunction with the accompanying drawings.

[0054] Figure 1 It is the logic block diagram of the industrial park environment brain platform of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] See also Figure 1 As shown, in one embodiment, an industrial park environmental brain platform is provided, and the platform includes a data acquisition end, a data processing end, an AI environmental brain and an analysis end, wherein the data acquisition end uses an existing water quality parameter monitor to achieve timed monitoring of the concentration of the detection factor of the drainage point, and the detection factors include common water quality pollutants and pollutants involved in the enterprises in the park. The specific monitoring process is not described in detail. It is set at each drainage point in the park, and then obtains the first sewage data of each drainage point at a fixed interval time point. The fixed time interval is set according to the selection and is not limited here. Then, the first sewage data is processed by the data processing end to obtain the second sewage data; then, the first sewage data and the second sewage data are traced and analyzed by the AI ​​environmental brain respectively to obtain the first tracing result and the second tracing result; the analysis end compares and judges the first tracing result and the second tracing result, and determines the selection of the tracing result according to the judgment and analysis result, and then obtains the pollution tracing result of the park. Through this process, the characteristics of the first sewage data and the second sewage data can be combined, that is, the accuracy of the tracing result is guaranteed, and the comprehensiveness of the judgment result is guaranteed.

[0057] As an implementation mode of the present invention, the process of processing the first sewage data includes: first, when it is monitored that the concentration of any detection factor in the first sewage data exceeds the threshold, the corresponding time point is used as the starting point of the tracing period time point, so as to start early warning of environmental risks, and then the concentration value of each detection factor in the first sewage data in the tracing period is obtained and fitted to obtain the concentration change curve of each detection factor. The specific fitting process can be implemented by the linear regression method in the prior art, which is not limited here, and then by the formula:

[0058]

[0059] Calculate the concentration value C of the i-th detection factor at the j-th time point i (t j ), the concentration values ​​of all detection factors are taken as the second sewage data; where i is the order of detection factor types, j is the order of fixed interval time points, C i (t) is the concentration curve fitted by the i-th detection factor, t j is the jth time point, Δt is a preset fixed period, which is selected and set according to empirical data. Therefore, through the above data processing process, on the one hand, the processed data can eliminate the influence of abnormal data on the judgment accuracy, and at the same time, based on the concentration value C i (t j ) can be used for analysis, and can make a comprehensive judgment based on the comprehensive situation of the changing trend at that time point on the basis of single-point data, thereby improving the reliability of the judgment result and further improving the accuracy of the traceability result.

[0060] As an implementation mode of the present invention, the first traceability result includes a first risk enterprise set R1, and the second traceability result includes a second risk enterprise set R2; the comparison, judgment and analysis process includes:

[0061] If R1=R2, it means that the results of the two judgments are consistent, so the accuracy of the judgment result is high. Therefore, all enterprises in the first risk enterprise set R1 and the second risk enterprise set R2 are judged as the pollution source tracing results of the park;

[0062] If R1≠R2, it means that there is a difference in the results of the two judgments. This difference may be caused by the elimination of error data or by the limitation of the data at a single time point. Therefore, firstly, the enterprise corresponding to R1∩R2 is judged as the result of the pollution source tracing in the park, and all enterprises in the sets R1-R2 and R2-R1 are matched and judged. According to the judgment results, it is determined whether to use the enterprises in the sets R1-R2 and R2-R1 as the results of the pollution source tracing in the park, and then the factors of both can be comprehensively considered to make judgments and selections to ensure the accuracy of the selection results.

[0063] As an implementation manner of the present invention, the matching judgment process includes:

[0064] Obtain the pollutant sequences of the untreated sewage of enterprises in the set R1-R2 and the set R2-R1 respectively;

[0065] By formula:

[0066]

[0067] Calculate the matching degree G between the x-th enterprise and the current test result x ;

[0068] Among them, M x The number of enterprises selected according to the relative emission concentration ranking of the detection factor for the xth enterprise, y∈[1, M x ], sort xy (t j ) is the yth detection factor of the xth enterprise at t j The relative concentration ranking of time points, f s is a comparison table function, which is a decreasing function, and sort xy (t j )>M x When f s (sort xy (t j ))=0; The comparison table function is formulated based on empirical data. The higher the ranking, the greater the result corresponding to the comparison table function, that is, the higher the corresponding matching degree. For all time points f s (sort xy (t j ), which combines the sequence matching of all time points for overall judgment, C yx (t j ) is the yth detection factor of the xth enterprise at t j The value at the time point, C0 y is the reference value of the yth detection factor, Ct y is the standard value of the yth detection factor, C0 y and Ct y are fixed values ​​for reference obtained from empirical data and satisfy Ct y >C0 y , so the obtained It can reflect the out-of-tolerance state of the detection factor. In addition, G1 is the matching threshold, which is selected after fitting based on empirical data. x The comparison process with the matching threshold G1, at the matching degree G xWhen ≥G1, the x-th enterprise is used as the pollution source tracing result of the park, thereby completing the matching judgment process more accurately.

[0069] As an embodiment of the present invention, the process of processing the first sewage data further includes: obtaining the slope k of the concentration curve fitted by the i-th detection factor at each time point ij ; At the same time, the analysis end is also used to calculate the slope k j , concentration value C i (t j ) and the corresponding time point to execute the early warning strategy. Through the prediction strategy, it can issue an early warning at a more adaptive time point according to the actual situation of water quality detection risks. Compared with the method of immediately issuing an early warning at the early stage of abnormal water quality parameters, this method can greatly reduce the increase in the workload of management personnel due to data misjudgment, while ensuring the timeliness of the early warning, and thus making it easier to deal with the environmental problems that arise; the specific process of the early warning strategy is: through the formula:

[0070]

[0071] Calculate the warning value W(H); where γ is the adjustment coefficient, which is selected and set based on the empirical data fitting, C0 i is the reference value of the ith detection factor, Ct i is the standard value of the i-th detection factor, which is the same as Ct in the above technical solution y and C0 y Same, N is the number of detection factors, i∈[1,N]; α i is the weight coefficient of the ith detection factor, which is set according to the degree of influence of the detection factor on the environment, H is the number of time points, and H is a variable. According to the comparison between the warning value and the warning threshold W1, a warning can be issued after the corresponding number of time points H when W(H)≥W1. Obviously, this method can greatly reduce the problem of false warnings caused by misjudgment of a single data. At the same time, the calculation process of the warning value is comprehensively judged according to factors such as data deviation and change speed, so that the best time point can be selected for warning, which is convenient for management personnel to manage the process.

[0072] It should be noted that the calculation process of the warning threshold W1 is set after fitting based on the empirical data of the park environment, and will not be further described here.

[0073] As an implementation mode of the present invention, the early warning strategy further includes:

[0074] When W0≤W(H)<W1, a consistency judgment is made based on the status of each drainage point in the park, and whether to issue an early warning is determined based on the result of the consistency judgment;

[0075] Among them, W0 is the early warning risk threshold, which is obtained based on empirical data, and W0<W1, which is used as the basic standard for judging whether to issue an early warning, that is, when W(H)≥W1 is reached, a direct early warning is issued. When W0≤W(H)<W1, in order to judge whether the deviation is caused by the overall water quality factors or the impact caused by the emission status of enterprises in some regions, an analysis is performed through the consistency judgment process. The consistency judgment process includes:

[0076] By formula:

[0077]

[0078] Calculate the risk consistency coefficient s of all drainage points in the park W ; Where s1 is the consistency reference threshold, Q is the number of drainage points in the park, z∈[1,Q], W z (H) is the warning value of the zth drainage point at the Hth time point, is the mean of the warning values ​​of all drainage points, so the risk consistency coefficient s W It reflects the discreteness of each location in the park relative to the whole, so the risk consistency coefficient s W ≤s1, it means that the error is caused by the overall water quality, so no warning will be issued temporarily. W >s1, it means that the water quality in some areas of the park is abnormal. At this time, an early warning will be issued to ensure that the local environmental problems in the park can be solved in time.

[0079] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An industrial park environment brain platform, characterized in that: The platform includes: The data collection terminal is set at each drainage point in the park to obtain the first sewage data at fixed intervals of each drainage point; A data processing end, used for processing the first sewage data to obtain second sewage data; The AI ​​environment brain is used to perform source tracing analysis on the first sewage data and the second sewage data respectively to obtain a first source tracing result and a second source tracing result; The analysis end is used to obtain the pollution source tracing results of the industrial park based on the comparison and analysis of the first source tracing results and the second source tracing results.

2. The industrial park environment brain platform according to claim 1 is characterized in that: The process of processing the first sewage data includes: When the concentration of any detection factor in the first sewage data is monitored to exceed the threshold, the corresponding time point is used as the starting point of the tracing period; Obtain the concentration value of each detection factor in the first sewage data during the tracing period and perform fitting to obtain the concentration change curve of each detection factor; By formula: Calculate the concentration value C of the i-th detection factor at the j-th time point i (t j ), the concentration values ​​of all detection factors are taken as the second sewage data; Among them, i is the order of the detection factor types, j is the order of the fixed interval time points, C i (t) is the concentration curve fitted by the i-th detection factor, t j is the jth time point, and Δt is a preset fixed period.

3. The industrial park environment brain platform according to claim 2 is characterized in that: The first tracing result includes a first venture enterprise set R1, and the second tracing result includes a second venture enterprise set R2; The process of comparison, judgment and analysis includes: If R1=R2, all enterprises in the first risk enterprise set R1 and the second risk enterprise set R2 are judged as the pollution source tracing results of the park; If R1≠R2, the enterprise corresponding to R1∩R2 is judged as the pollution source tracing result of the industrial park, and all enterprises in the set R1-R2 and the set R2-R1 are matched and judged. According to the judgment result, it is determined whether to use the enterprises in the set R1-R2 and the set R2-R1 as the pollution source tracing result of the industrial park.

4. The industrial park environment brain platform according to claim 3 is characterized in that: The matching judgment process includes: Obtain the pollutant sequences of the untreated sewage of enterprises in the set R1-R2 and the set R2-R1 respectively; By formula: Calculate the matching degree G between the x-th enterprise and the current test result x ; If the matching degree G x ≥G1, the xth enterprise will be taken as the pollution source tracing result of the park; Among them, M x The number of enterprises selected according to the relative emission concentration ranking of the detection factor for the xth enterprise, y∈[1, M x ]; sort xy (t j ) is the yth detection factor of the xth enterprise at t j The relative concentration ranking of time points, f s is a comparison table function, which is a decreasing function, and sort xy (t j )>M x When f s (sort xy (t j ))=0; For all time points f s (sort xy (t j ))’s mean, C yx (t j ) is the yth detection factor of the xth enterprise at t j The value at the time point, C0 y is the reference value of the yth detection factor, Ct y is the standard value of the yth detection factor, and G1 is the matching threshold.

5. The industrial park environment brain platform according to claim 2 is characterized in that: The process of processing the first sewage data also includes: Get the slope k of the concentration curve fitted by the i-th detection factor at each time point ij ; The analysis end is also used to calculate the slope k j , concentration value C i (t j ) and execute early warning strategies at corresponding time points.

6. The industrial park environment brain platform according to claim 5 is characterized in that: The early warning strategy includes: By formula: Calculate and obtain the warning value W(H); When the warning value W(H)≥W1, a warning is issued; Among them, W1 is the warning threshold, γ is the adjustment coefficient, C0 i is the reference value of the ith detection factor, Ct i is the standard value of the i-th detection factor, N is the number of detection factors, i∈[1,N]; α i is the weight coefficient of the i-th detection factor, H is the number of time points, and H is the variable.

7. The industrial park environment brain platform according to claim 6 is characterized in that: The early warning strategy also includes: When W0≤W(H)<W1, a consistency judgment is made based on the status of each drainage point in the park, and whether to issue an early warning is determined based on the result of the consistency judgment; Among them, W0 is the warning risk threshold.

8. The industrial park environment brain platform according to claim 7 is characterized in that: The consistency judgment process includes: By formula: Calculate the risk consistency coefficient s of all drainage points in the park W ; If the risk consistency coefficient s W ≤s1, no warning will be issued; If W >s1, an early warning is issued; Among them, s1 is the consistency reference threshold, Q is the number of drainage points in the park, z∈[1,Q], W z (H) is the warning value of the zth drainage point at the Hth time point, It is the average of the warning values ​​of all drainage points.

Images