A method and system for monitoring mixed flue gas in different regions

Abstract

The present invention relates to the field of data processing technology, and in particular to a regional mixed flue gas patrol monitoring method and system, comprising: dividing a flue gas duct into multiple partitions, obtaining smoke concentration data and flue gas flow rate data, and constructing a smoke concentration curve and a flue gas flow rate curve; obtaining a patrol curve based on the patrol sorting of the smoke concentration data, and performing window division; obtaining the similarity of each smoke concentration data of all partitions in each smoke concentration window, combining the patrol data of the patrol window, obtaining the smoke aggregation degree of each patrol window, combining the possibility of smoke pollution abnormality of all partitions in each smoke flow rate window, obtaining the smoke pollution degree of all partitions in each smoke concentration window, performing back-purge, comparing the difference in smoke pollution degree before and after back-purge, obtaining smoke emission abnormality evaluation, and implementing smoke early warning. The present invention aims to improve the accuracy of smoke abnormality early warning in the process of regional mixed flue gas patrol monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for monitoring mixed flue gas in different regions. Background Art

[0002] The continuous monitoring system for flue gas emissions is used to monitor the emission of pollutants such as smoke particles and nitrogen oxides in order to determine the cost of pollution prevention and control for enterprises. Usually, the continuous monitoring system for flue gas emissions includes temperature, pressure, flow rate sensors and dust monitors, etc., which can detect and warn of atmospheric greenhouse gas emissions in real time. Through the continuous monitoring system for flue gas emissions and smoke concentration sensors, the flue pollutant emissions can be accurately monitored.

[0003] During the collection of smoke concentration data, dust accumulation occurs, which affects the measurement accuracy of humidity, temperature, gas flow rate and other data, and further affects the accuracy of pollutant warning. Existing methods usually perform backflush operations by setting fixed time intervals. During the backflush process, in addition to detecting whether the pressure of the pressure source is normal, it also plays a role in regularly cleaning the accumulated dust to reduce the impact of dust accumulation. However, the time for smoke accumulation to reach the threshold is not fixed, resulting in the backflush operation may be performed when it is less than the threshold time, thereby increasing the extra workload of the compressor and increasing energy consumption; if the smoke accumulation rate is fast, the smoke data may have exceeded the normal range before the set backflush threshold time is reached, causing the system to misjudge that the smoke data is abnormal and needs to be warned when there is no backflush operation, affecting the accuracy of the warning. Summary of the invention

[0004] The present invention provides a mixed smoke patrol monitoring method and system for different regions to solve the existing problems.

[0005] A regional mixed flue gas patrol monitoring method and system of the present invention adopts the following technical solutions:

[0006] An embodiment of the present invention provides a method for monitoring mixed flue gas in different regions, the method comprising the following steps:

[0007] Divide the flue gas duct into multiple zones, obtain smoke concentration data and flue gas flow rate data of each zone, and construct a smoke concentration curve and a flue gas flow rate curve for each zone;

[0008] The smoke concentration data of the smoke concentration curves of all partitions are sorted based on the patrol, and the patrol curve and the smoke concentration curve of each partition are divided into windows to obtain a number of patrol windows and smoke concentration windows; according to the concentration increment of the smoke concentration data in each smoke concentration window, the similarity of each smoke concentration data of all partitions in each smoke concentration window is obtained; according to the similarity of each smoke concentration data of all partitions in each smoke concentration window, and the patrol data of each patrol window in the patrol curve, the smoke aggregation degree of each patrol window is obtained;

[0009] Obtaining a flue gas flow rate window, and obtaining the possibility of abnormal smoke dust pollution in each flue gas flow rate window for all partitions according to the difference of flue gas flow rate data in each flue gas flow rate window in the flue gas flow rate curve of each partition;

[0010] According to the possibility of smoke pollution abnormality in the flue gas flow rate window and the smoke accumulation degree in the patrol window corresponding to the flue gas flow rate window, the smoke pollution degree of all partitions in each smoke concentration window is obtained; according to the smoke pollution degree of all partitions in each smoke concentration window, back-purge is performed to obtain the smoke pollution degree of all partitions after back-purge;

[0011] According to the difference in smoke pollution degree before and after back-purge, a smoke emission abnormality evaluation is obtained, and a smoke early warning is issued according to the smoke emission abnormality evaluation.

[0012] Furthermore, the specific method of obtaining the patrol curve is:

[0013] Take every 1 second as a moment, and obtain the smoke concentration data and smoke flow velocity data of each partition at each moment through the smoke concentration sensors and smoke flow velocity sensors of all partitions. Each sensor collects the smoke concentration data and smoke flow velocity data of all partitions once at each moment;

[0014] The smoke concentration data of the first partition at the first moment is used as the first sequence data in the patrol data sequence, and the smoke concentration data of the next partition at the next moment is used as the next sequence data in the patrol data sequence, wherein the next partition of the fourth partition is the first partition, and so on, to construct the patrol data sequence;

[0015] The patrol curve is obtained by taking the collection time of the patrol data sequence as the horizontal coordinate and the data corresponding to the collection time as the vertical coordinate.

[0016] Furthermore, the specific method of obtaining the plurality of patrol windows and smoke concentration windows is as follows:

[0017] Sliding segmentation of the smoke concentration curve of each partition is performed using a preset time window to obtain a number of smoke concentration windows for each partition;

[0018] The patrol curve is segmented by sliding using a preset time window to obtain a number of patrol windows.

[0019] Furthermore, obtaining the similarity of each smoke concentration data in each smoke concentration window of all partitions includes:

[0020] The calculation formula for the similarity of the i-th smoke concentration data in the m-th smoke concentration window of all partitions is:

[0021]

[0022] Among them, G m,i represents the similarity of the ith smoke concentration data in the mth smoke concentration window of all partitions; d j,m,i It represents the increment of the ith smoke concentration data in the mth smoke concentration window in the jth partition; J represents the number of partitions; exp() represents an exponential function with a natural constant as the base; || represents the absolute value function.

[0023] Furthermore, obtaining the smoke and dust accumulation degree of each patrol window includes:

[0024] The calculation formula for the smoke and dust accumulation degree of the mth patrol window in the patrol curve is:

[0025]

[0026] Among them, G m represents the dust accumulation degree of the mth patrol window in the patrol curve; a represents the number of patrol data in the patrol window; G m,i represents the similarity of the ith smoke concentration data in the mth smoke concentration window of all partitions; r i represents the ith patrol data in the patrol curve; r represents the preset smoke concentration threshold; exp() represents an exponential function with a natural constant as the base.

[0027] Furthermore, the possibility of obtaining smoke pollution anomalies in all partitions in each smoke flow rate window includes:

[0028] Using a preset time window, the flue gas flow rate curve of each partition is subjected to sliding segmentation to obtain a plurality of flue gas flow rate windows of each partition;

[0029] The calculation formula for the probability of smoke pollution anomaly in the mth window of all partitions is:

[0030]

[0031] Among them, P m Indicates the possibility of abnormal smoke pollution in the mth flue gas flow rate window in the flue gas flow rate curve of all partitions; bm represents the number of flue gas velocity data in the mth window; v m,k represents the variance of the kth flue gas flow rate data in the mth flue gas flow rate window in the flue gas flow rate curve of each partition; u represents the number of flue gas flow rate increments greater than zero in the flue gas flow rate curves of all partitions, and the flue gas flow rate increment is obtained by recording the difference between the flue gas flow rate data in the flue gas flow rate curves of all partitions and the previous flue gas flow rate data as the flue gas flow rate increment; τ m It represents the discrete degree of flue gas flow rate in the mth flue gas flow rate window in the flue gas flow rate curve of all partitions; norm[] represents the linear normalization function.

[0032] Furthermore, the specific steps for obtaining the discrete degree of the flue gas flow rate within the flue gas flow rate window are:

[0033] The sum of the differences between the mean of each flue gas flow rate data in the mth window in the flue gas flow rate curves of all partitions and the mean of the previous flue gas flow rate data is recorded as the discrete degree of the flue gas flow rate in the mth flue gas flow rate window in the flue gas flow rate curves of all partitions.

[0034] Furthermore, obtaining the smoke pollution degree after back-purging of all partitions includes:

[0035] The calculation formula for the smoke pollution degree of the mth smoke concentration window of all partitions is:

[0036]

[0037] Among them, w m Indicates the smoke pollution degree of the mth smoke concentration window of all partitions; P m Indicates the possibility of abnormal smoke pollution in the mth flue gas flow rate window in the flue gas flow rate curve of all partitions; G m Indicates the degree of smoke and dust accumulation in the mth patrol window in the patrol curve; Represents preset hyperparameters;

[0038] When the smoke pollution degree is greater than or equal to the preset pollution degree threshold, back-purging is performed to obtain the smoke pollution degrees of all partitions after back-purging.

[0039] Furthermore, obtaining the smoke emission abnormality evaluation includes:

[0040] The calculation formula for the abnormal evaluation of smoke emissions is:

[0041]

[0042] Where Q represents the abnormal evaluation of smoke emission; w m ′ represents the smoke pollution degree before back-purge of the mth smoke concentration window in all partitions; wm ″ represents the smoke pollution degree of the mth smoke concentration window of all partitions after back-purging; M represents the number of smoke concentration windows of the partition; || represents the absolute value function; exp() represents the exponential function with a natural constant as the base.

[0043] The present invention also proposes a regional mixed smoke patrol monitoring system, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program stored in the memory to implement the steps of the aforementioned regional mixed smoke patrol monitoring method.

[0044] The beneficial effects of the technical solution of the present invention are as follows: dividing the flue gas duct into multiple partitions, obtaining smoke concentration data and smoke flow rate data of each partition, constructing a smoke concentration curve and a smoke flow rate curve of each partition, obtaining a patrol curve based on the patrol sorting of the smoke concentration data of the smoke concentration curves of all partitions, which is used to reflect the eddy effect of smoke in the flue; performing window division on the patrol curve and the smoke concentration curve of each partition to obtain a plurality of patrol windows and smoke concentration windows, obtaining the similarity of each smoke concentration data of all partitions in each smoke concentration window according to the concentration increment of the smoke concentration data in each smoke concentration window, obtaining the degree of smoke aggregation of each patrol window according to the similarity of each smoke concentration data of all partitions in each smoke concentration window and the patrol data of each patrol window in the patrol curve, in order to analyze the actual concentration of smoke in the flue; obtaining the smoke flow rate window, and The difference in flue gas velocity data in each flue gas velocity window in the gas velocity curve is used to obtain the possibility of abnormal smoke pollution in all partitions in each flue gas velocity window, reflecting the probability of whether smoke is easy to gather due to flue gas velocity; the degree of smoke pollution in all partitions in each smoke concentration window is obtained according to the possibility of abnormal smoke pollution in the flue gas velocity window and the degree of smoke aggregation in the patrol window corresponding to the flue gas velocity window; back-purge is performed according to the smoke pollution degree in each smoke concentration window of all partitions to obtain the smoke pollution degree after back-purge in all partitions, and the difference in smoke concentration before and after back-purge is analyzed to determine the abnormal situation of smoke emission, providing a basis and support for accurate smoke warning; based on the difference in smoke pollution degree before and after back-purge, the evaluation of smoke emission abnormality is obtained, and smoke warning is carried out based on the evaluation of smoke emission abnormality, so as to improve the accuracy of smoke abnormality warning during the patrol monitoring of mixed flue gas in different regions. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 The present invention is a flowchart of the steps of a method for monitoring mixed flue gas in different regions. DETAILED DESCRIPTION

[0047] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a regional mixed flue gas patrol monitoring method and system proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0048] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0049] The specific scheme of the regional mixed flue gas patrol monitoring method and system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0050] See also Figure 1 , which shows a flowchart of a method for monitoring mixed flue gas in different regions according to an embodiment of the present invention, the method comprising the following steps:

[0051] Step S001: Divide the flue gas duct into multiple partitions, obtain smoke concentration data and smoke flow rate data of each partition, and construct a smoke concentration curve and a smoke flow rate curve of each partition.

[0052] A sensor is placed inside the pipeline counterclockwise every 45 degrees along the edge of the circular cross-section of the pipeline, where smoke concentration sensors and flue gas flow velocity sensors are installed alternately, and each group of smoke concentration sensors and flue gas flow velocity sensors is used as a collection partition. The first partition, the second partition, the third partition and the fourth partition are obtained according to the installation order of the sensors. Every 1 second is used as a moment, and the smoke concentration data and flue gas flow velocity data of each partition at each moment are obtained through the smoke concentration sensors and flue gas flow velocity sensors of all partitions. Each sensor collects the smoke concentration data and flue gas flow velocity data of all partitions once at each moment, and the collection time of the smoke concentration data and the flue gas flow velocity data is used as the horizontal coordinate, and the data corresponding to the collection time is used as the vertical coordinate to obtain the smoke concentration curve and the flue gas flow velocity curve.

[0053] Step S002: Obtain a patrol curve based on patrol sorting of the smoke concentration data of the smoke concentration curves of all partitions, divide the patrol curve and the smoke concentration curve of each partition into windows, and obtain a number of patrol windows and smoke concentration windows; obtain the similarity of each smoke concentration data in each smoke concentration window of all partitions according to the concentration increment of the smoke concentration data in each smoke concentration window; obtain the smoke aggregation degree of each patrol window according to the similarity of each smoke concentration data in each smoke concentration window of all partitions and the patrol data of each patrol window in the patrol curve.

[0054] It should be noted that pollutant emissions often contain a large amount of smoke and dust. Flue gas usually contains resistance structures such as elbows and reducers, which are prone to eddy currents. As a result, the smoke concentrations at different positions of the same cross section may be different. At the same time, the distribution of flue gas velocity will also be different. Slower flue gas velocity makes it easier for larger particles to adhere, resulting in lower local smoke concentrations; while faster flue gas velocity makes it easier for larger particles to suspend, resulting in higher local smoke concentrations. In addition, large differences in concentration data in different areas may indicate that the sensor surface is covered with pollutants and smoke is seriously accumulated in the flue gas. Therefore, according to the eddy current effect presented by the smoke concentration data of different partitions in the flue, this embodiment re-sorts the smoke concentration data of the smoke concentration curves of all partitions to obtain a patrol curve, which is used to reflect the eddy current effect of smoke in the flue.

[0055] Specifically, the smoke concentration data of the first partition at the first moment is used as the first sequence data in the patrol data sequence, the smoke concentration data of the next partition at the next moment is used as the next sequence data in the patrol data sequence, and so on, to construct the patrol data sequence; it should be noted that, since the flue is a closed circle, the next partition of the fourth partition is the first partition. As an example, the smoke concentration data of the first partition at the first moment is used as the first sequence data in the patrol data sequence, the smoke concentration data of the second partition at the second moment is used as the second sequence data in the patrol data sequence, the smoke concentration data of the third partition at the third moment is used as the third sequence data in the patrol data sequence, the smoke concentration data of the third partition at the fourth moment is used as the fourth sequence data in the patrol data sequence, the smoke concentration data of the first partition at the fifth moment is used as the fifth sequence data in the patrol data sequence, and so on, until the patrol data sequence is obtained; further, the patrol curve is obtained according to the method for constructing the smoke concentration curve.

[0056] It should be further explained that when collecting smoke concentration in the flue, the data at different locations can be fully obtained by combining sampling at multiple sampling points in a patrol manner. Due to the existence of the resistance structure, the concentration performance of different partitions may be inconsistent, that is, the greater the difference in the smoke concentration curve of each partition, and the greater the deviation from the standard empirical value, the higher the degree of smoke aggregation. In order to analyze the true concentration of smoke in the flue, each patrol data in the patrol curve comes from the smoke concentration data sequence of each partition. Therefore, each patrol data in the patrol curve has similarities between all partitions. The degree of smoke aggregation is obtained by the difference between the smoke concentration curve of each partition and the patrol curve.

[0057] Specifically, a time window is preset. The length of the time window in this embodiment is described using 3 seconds as an example. Other embodiments may set other values, which are not specifically limited in this embodiment. The time window is used to slide and divide the smoke concentration curve of each partition to obtain several smoke concentration windows for each partition.

[0058] In each smoke concentration window in each partition, the difference between each smoke concentration data and the previous smoke concentration data is recorded as the increment of each smoke concentration data in each smoke concentration window in each partition. The calculation formula for the similarity of the i-th smoke concentration data in the m-th smoke concentration window in all partitions is:

[0059]

[0060] Among them, G m,i represents the similarity of the ith smoke concentration data in the mth smoke concentration window of all partitions; d j,m,iIt represents the increment of the ith smoke concentration data in the mth smoke concentration window in the jth partition; J represents the number of partitions; exp() represents an exponential function with a natural constant as the base; || represents the absolute value function.

[0061] It represents the absolute value of the product of the increments of the ith smoke concentration data of all partitions in the mth smoke concentration window. The value reflects the overall change of all partitions relative to the previous data in the window. The larger the value, the more consistent the concentration change trend of all partitions and the lower the degree of smoke aggregation. It represents the product of the square roots of the increments of the ith smoke concentration data of all partitions in the mth smoke concentration window. The value represents the magnitude of the overall change of all partitions in the window relative to the previous data. The larger the value, the more consistent the data change trend of all partitions and the higher the degree of smoke aggregation.

[0062] It should be noted that each patrol data in the patrol curve has similarity between all partitions. If the similarity of a patrol data is higher and lower than the preset smoke concentration threshold, the performance is more normal and the smoke concentration is lower. On the contrary, if the similarity of a patrol data is lower and higher than the preset smoke concentration threshold, the smoke concentration is higher.

[0063] The patrol window is obtained according to the construction method of the smoke concentration window. The calculation formula of the smoke concentration degree of the mth patrol window in the patrol curve is:

[0064]

[0065] Among them, G m represents the dust accumulation degree of the mth patrol window in the patrol curve; a represents the number of patrol data in the patrol window; G m,i represents the similarity of the ith smoke concentration data in the mth smoke concentration window of all partitions; r i Represents the ith patrol data in the patrol curve; r represents a preset smoke concentration threshold value. In this embodiment, the preset smoke concentration threshold value r=2 mg / m3 is used as an example for description. Other implementation methods may set it to other values, which are not limited in this embodiment; exp() represents an exponential function with a natural constant as the base.

[0066] exp(r i ―r) represents the exponential function between the ith patrol data in the patrol curve and the preset smoke concentration threshold. The larger the value, the greater the difference between a patrol data in the patrol curve and the preset smoke concentration threshold. Indicates the degree of aggregation of the ith patrol data in the mth window in the patrol curve. It represents the inverse of similarity, that is, the degree of aggregation of the patrol data, and multiplying it by exp(r i ―r) takes into account the difference between the patrol data and the preset smoke concentration threshold. The larger the value, the higher the degree of aggregation of the patrol data.

[0067] Step S003: Obtain a flue gas flow rate window, and obtain the possibility of smoke dust pollution abnormality in each flue gas flow rate window for all partitions according to the difference of flue gas flow rate data in each flue gas flow rate window in the flue gas flow rate curve of each partition.

[0068] It should be noted that different air flow rates have different effects on the degree of smoke accumulation. The faster the air flow rate, the easier it is for suspended particles to remain suspended, so the measured smoke concentration is higher. In each partition, an air flow rate sensor is installed. If the difference in flow rate data between different partitions is greater, and the growth rate change is more obvious and faster, it means that the possibility of smoke accumulation is higher.

[0069] Specifically, the smoke flow rate window is obtained according to the construction method of the smoke concentration window. The calculation formula for the probability of smoke pollution abnormality in the mth window of all partitions is:

[0070]

[0071] Among them, P m Indicates the possibility of abnormal smoke pollution in the mth flue gas flow rate window in the flue gas flow rate curve of all partitions; b m represents the number of flue gas velocity data in the mth window; v m,k represents the variance of the kth flue gas flow rate data in the mth flue gas flow rate window in the flue gas flow rate curve of each partition; u represents the number of flue gas flow rate increments greater than zero in the flue gas flow rate curves of all partitions, and the flue gas flow rate increment is obtained by recording the difference between the flue gas flow rate data in the flue gas flow rate curves of all partitions and the previous flue gas flow rate data as the flue gas flow rate increment; τ m It represents the discrete degree of flue gas flow rate in the mth flue gas flow rate window in the flue gas flow rate curves of all partitions. The discrete degree of flue gas flow rate is obtained by taking the sum of the differences between the mean of each flue gas flow rate data in the mth window in the flue gas flow rate curves of all partitions and the mean of the previous flue gas flow rate data as the discrete degree of flue gas flow rate in the mth flue gas flow rate window in the flue gas flow rate curves of all partitions. norm[] represents a linear normalization function, which normalizes the data values ​​to the interval [0,1].

[0072] It represents the average value of the discrete degree of the flue gas flow rate data in each window. The larger the value is, the greater the discrete degree of the flue gas flow rate data in the window is. It reflects the degree of dispersion of the flue gas velocity in the window and the number of flue gas velocity increments greater than zero. The size of these values ​​reflects the possibility of abnormal smoke pollution. If these values ​​are large, it means that the possibility of smoke pollution is higher.

[0073] Step S004: According to the possibility of abnormal smoke pollution in the flue gas flow rate window and the smoke concentration degree in the patrol window corresponding to the flue gas flow rate window, the smoke pollution degree of all partitions in each smoke concentration window is obtained; according to the smoke pollution degree of all partitions in each smoke concentration window, back-purge is performed to obtain the smoke pollution degree of all partitions after back-purge.

[0074] It should be noted that the possibility of abnormal smoke pollution reflects the probability of whether smoke is easy to gather due to the smoke flow rate, and the degree of smoke gathering indicates the specific actual situation of smoke gathering. Therefore, if the possibility of smoke gathering is greater and the degree of smoke gathering is smaller, it means that the degree of smoke pollution is smaller. On the contrary, if the possibility of smoke gathering is smaller and the degree of smoke gathering is greater, it means that the degree of smoke pollution is greater.

[0075] The calculation formula for the smoke pollution degree of the mth smoke concentration window of all partitions is:

[0076]

[0077] Among them, w m Indicates the smoke pollution degree of the mth smoke concentration window of all partitions; P m Indicates the possibility of abnormal smoke pollution in the mth flue gas flow rate window in the flue gas flow rate curve of all partitions; G m Indicates the degree of smoke and dust accumulation in the mth patrol window in the patrol curve; Represents a preset hyperparameter to avoid the denominator being 0.

[0078] The pollution level threshold value c=0.7 preset in this embodiment is used as an example for description. Other implementations may be set to other values, which are not limited in this embodiment. When the smoke pollution level is greater than or equal to the preset pollution level threshold value c, after back-purge, the purge time is adaptively realized, and the smoke pollution level after back-purge of all partitions is obtained according to the above steps; when the smoke pollution level is less than the preset pollution level threshold value c, back-purge is not performed.

[0079] Step S005: obtaining a smoke emission abnormality evaluation based on the difference in smoke pollution degree before and after the back-purge, and issuing a smoke early warning based on the smoke emission abnormality evaluation.

[0080] It should be noted that the purge time of the flue gas emission continuous monitoring system can remove the smoke accumulated near each sensor and detect whether the bottom pressure source is normal. In the process of removing smoke, the possibility of real-time smoke abnormality should gradually decrease, and the difference in possibility before and after purge should be large. If after the back-purge, the real-time pollution degree is less different from the real-time pollution degree before the back-purge, it means that the air pollution level is relatively serious, and it is necessary to determine whether to issue an early warning based on the threshold.

[0081] The calculation formula for the abnormal evaluation of smoke emissions is:

[0082]

[0083] Where Q represents the abnormal evaluation of smoke emission; w m ′ represents the smoke pollution degree before back-purge of the mth smoke concentration window in all partitions; w m ″ represents the smoke pollution degree of the mth smoke concentration window of all partitions after back-purging; M represents the number of smoke concentration windows of the partition; || represents the absolute value function; exp() represents the exponential function with a natural constant as the base.

[0084] |w m ′―w m ″| represents the absolute value of the difference in smoke pollution degree before and after back-purging in the mth smoke concentration window of all partitions. This value represents the change in smoke concentration before and after back-purging in each partition. The larger the value, the greater the difference in smoke concentration before and after back-purging, that is, the greater the change in smoke. It means summing up the smoke concentration windows of all partitions. This value reflects the overall change of smoke concentration in all partitions. The larger the value, the greater the change of the overall smoke concentration, indicating that the abnormal degree of smoke emission increases, that is, the abnormal degree of smoke emission or the drastic degree of change is higher.

[0085] The preset abnormal threshold g=0.6 in this embodiment is used as an example for description. Other implementations may be set to other values, which are not limited in this embodiment. When the smoke emission abnormality evaluation is less than the preset abnormal threshold g, there is no need to issue a smoke emission warning; when the smoke emission abnormality evaluation is greater than or equal to the preset abnormal threshold g, a smoke emission warning is issued to achieve the purpose of real-time warning.

[0086] So far, the present invention is completed.

[0087] To sum up, in an embodiment of the present invention, the present invention also provides a regional mixed smoke patrol monitoring system, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program stored in the memory to implement the steps of the aforementioned regional mixed smoke patrol monitoring method.

[0088] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A regional mixed flue gas patrol monitoring method, characterized in that: The method comprises the following steps: Divide the flue gas duct into multiple zones, obtain smoke concentration data and flue gas flow rate data of each zone, and construct a smoke concentration curve and a flue gas flow rate curve for each zone; The smoke concentration data of the smoke concentration curves of all partitions are sorted based on the patrol, and the patrol curve and the smoke concentration curve of each partition are divided into windows to obtain a number of patrol windows and smoke concentration windows; according to the concentration increment of the smoke concentration data in each smoke concentration window, the similarity of each smoke concentration data of all partitions in each smoke concentration window is obtained; according to the similarity of each smoke concentration data of all partitions in each smoke concentration window, and the patrol data of each patrol window in the patrol curve, the smoke aggregation degree of each patrol window is obtained; Obtaining a flue gas flow rate window, and obtaining the possibility of abnormal smoke dust pollution in each flue gas flow rate window for all partitions according to the difference of flue gas flow rate data in each flue gas flow rate window in the flue gas flow rate curve of each partition; According to the possibility of smoke pollution abnormality in the flue gas flow rate window and the smoke accumulation degree in the patrol window corresponding to the flue gas flow rate window, the smoke pollution degree of all partitions in each smoke concentration window is obtained; according to the smoke pollution degree of all partitions in each smoke concentration window, back-purge is performed to obtain the smoke pollution degree of all partitions after back-purge; According to the difference in smoke pollution degree before and after back-purge, obtain smoke emission abnormality evaluation, and issue smoke early warning according to the smoke emission abnormality evaluation; The obtaining of the similarity of each smoke concentration data in each smoke concentration window of all partitions includes: All partitions in Within the smoke concentration window The calculation formula for the similarity of smoke concentration data is: in, Indicates that all partitions are Within the smoke concentration window The similarity of smoke concentration data; Indicates Partition Within the smoke concentration window The increment of smoke concentration data; Indicates the number of partitions; represents an exponential function with a natural constant as base; represents the absolute value function; The step of obtaining the smoke and dust accumulation degree of each patrol window includes: The first The calculation formula for the smoke and dust accumulation degree of each patrol window is: in, Indicates the first The degree of smoke and dust accumulation at each patrol window; Indicates the number of patrol data in the patrol window; Indicates that all partitions are Within the smoke concentration window The similarity of smoke concentration data; Indicates the first patrol data; Indicates the preset smoke concentration threshold; represents an exponential function with a natural constant as base; The possibility of obtaining smoke pollution anomalies in each smoke flow rate window for all partitions includes: Using a preset time window, the flue gas flow rate curve of each partition is subjected to sliding segmentation to obtain a plurality of flue gas flow rate windows of each partition; All partitions The calculation formula for the probability of abnormal smoke pollution in each window is: in, Indicates the flue gas velocity curve of all zones. The possibility of abnormal smoke pollution in each flue gas flow rate window; Indicates The number of flue gas velocity data in a window; Indicates the flue gas velocity curve of each zone Within the flue gas velocity window The variance of the flue gas velocity data; Indicates the number of flue gas flow rate increments greater than zero in the flue gas flow rate curves of all partitions, wherein the flue gas flow rate increment is obtained by recording the difference between the flue gas flow rate data in the flue gas flow rate curves of all partitions and the previous flue gas flow rate data as the flue gas flow rate increment; Indicates the flue gas velocity curve of all zones. The degree of dispersion of flue gas flow rate within a flue gas flow rate window; represents the linear normalization function.

2. According to claim 1, a method for monitoring mixed flue gas in different regions is characterized in that: The specific method of obtaining the patrol curve is as follows: Take every 1 second as a moment, and obtain the smoke concentration data and smoke flow velocity data of each partition at each moment through the smoke concentration sensors and smoke flow velocity sensors of all partitions. Each sensor collects the smoke concentration data and smoke flow velocity data of all partitions once at each moment; The smoke concentration data of the first partition at the first moment is used as the first sequence data in the patrol data sequence, and the smoke concentration data of the next partition at the next moment is used as the next sequence data in the patrol data sequence, wherein the next partition of the fourth partition is the first partition, and so on, to construct the patrol data sequence; The patrol curve is obtained by taking the collection time of the patrol data sequence as the horizontal coordinate and the data corresponding to the collection time as the vertical coordinate.

3. According to claim 1, a method for monitoring mixed flue gas in different regions is characterized in that: The specific method of obtaining the plurality of patrol windows and smoke concentration windows is as follows: Sliding segmentation of the smoke concentration curve of each partition is performed using a preset time window to obtain a number of smoke concentration windows for each partition; The patrol curve is segmented by sliding using a preset time window to obtain a number of patrol windows.

4. According to claim 1, a regional mixed smoke patrol monitoring method is characterized in that: The specific steps for obtaining the discrete degree of the flue gas flow rate within the flue gas flow rate window are as follows: The flue gas velocity curves of all zones The sum of the differences between the mean of each flue gas flow rate data in a window and the mean of the previous flue gas flow rate data is recorded as the flue gas flow rate curve of all partitions. The degree of discreteness of flue gas flow rate within a flue gas flow rate window.

5. The method for monitoring mixed flue gas in different regions according to claim 1, characterized in that: The method of obtaining the smoke pollution degree after back-purging of all partitions includes: All partitions The calculation formula for the smoke pollution degree of a smoke concentration window is: in, Indicates the first The degree of smoke pollution in each smoke concentration window; Indicates the flue gas velocity curve of all zones. The possibility of abnormal smoke pollution in each flue gas flow rate window; Indicates the first The degree of smoke and dust accumulation at each patrol window; Represents preset hyperparameters; When the smoke pollution degree is greater than or equal to the preset pollution degree threshold, back-purging is performed to obtain the smoke pollution degrees of all partitions after back-purging.

6. A regional mixed smoke patrol monitoring method according to claim 1, characterized in that: The obtaining of smoke emission abnormality evaluation comprises: The calculation formula for the abnormal evaluation of smoke emissions is: in, It indicates abnormal evaluation of smoke emission; Indicates the first The degree of smoke pollution before back-purge in each smoke concentration window; Indicates the first The degree of smoke pollution after back-purge in each smoke concentration window; Indicates the number of smoke concentration windows in the partition; represents the absolute value function; Represents an exponential function with a natural constant as its base.

7. A regional mixed smoke patrol monitoring system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, a method for monitoring mixed flue gas in different regions according to any one of claims 1 to 6 is implemented.

Images