A big data analysis method to identify compliance before and after operation and maintenance

Abstract

The present invention relates to a big data analysis method for identifying compliance before and after operation and maintenance, and belongs to the field of electrical digital data processing. The method is applied to an environmental monitoring station, comprising: using a first big data analysis node to output a deep neural network after completing multiple learnings as an intelligent emission analysis model; using a second big data analysis node to intelligently analyze the predicted value of the time required for the emission gas of the thermal power unit to meet the environmental pollution standard after detecting that the thermal power unit of the target power plant has been shut down for unit operation and maintenance, and using the intelligent emission analysis model; by using a customized artificial intelligence model to intelligently analyze the predicted value of the time required for the emission gas of the thermal power unit of the target power plant to meet the compliance after the thermal power unit of the target power plant has been shut down for unit operation and maintenance based on multiple basic information screened in a targeted manner, the environmental monitoring station effectively and timely analyzes the collected emission data to achieve carbon emission monitoring.

Description

Technical Field

[0001] The present invention relates to the field of electronic digital data processing, and in particular to a big data analysis method for identifying compliance before and after operation and maintenance. Background Art

[0002] In modern industrial production, especially in high-pollution industries, changes in emission concentration are often affected by a variety of factors. Among them, the operation and maintenance of production equipment is one of the important reasons for sudden changes in emission concentration. With the continuous improvement of environmental protection standards, enterprises urgently need to monitor changes in emission concentration in real time so that they can quickly identify and correct emission anomalies caused by operation and maintenance errors or equipment failures such as parameter configuration deviations. Although existing emission monitoring systems can provide real-time data on emission concentrations, there are still some significant problems such as limited data processing capabilities, untimely tracking of pollution information, lack of personalized analysis, and lack of configuration parameter tuning.

[0003] For example, the Chinese invention patent with publication number CN104457846A proposes a flue gas emission monitoring method, which relates to the field of coal-fired technology and achieves the purpose of automatically realizing real-time monitoring of flue gas emission data as a whole. The main technical scheme of the present invention is: after the master control center server establishes a network connection with multiple sub-center data servers, the master control center server obtains the collected and / or online power data and the identification information of each sub-center data server through the network connection; the master control center server analyzes the effectiveness, timeliness and rationality of the flue gas emission data and / or online power data to obtain reasonable and effective flue gas emission data and / or online power data; the master control center server outputs and displays the reasonable and effective flue gas emission data and / or online power data of each sub-center data server according to the identification information, so as to realize online monitoring of flue gas emissions. The present invention is mainly used for monitoring flue gas emissions of thermal power units in power plants.

[0004] For example, the Chinese invention patent with publication number CN106731407A proposes an ultra-low emission system and process for purifying flue gas from carbon anode roasting. The system includes a spray cooler, an electrostatic tar collector, and a desulfurization and defluorination device. The flue gas output end of the spray cooler is connected to the flue gas inlet of the electrostatic tar collector, and the flue gas outlet of the electrostatic tar collector is connected to the flue gas inlet of the desulfurization and defluorination device through a smoke exhaust fan; the anode system in the electrostatic tar collector adopts a combined multifunctional double-peak plate anode row, and the cathode system adopts a prismatic line and a square line; a hot water cleaning device is provided on the electrostatic tar collector. The various pollutant emission indicators of the present invention can meet the requirements for atmospheric pollutant emissions. The flue gas purification efficiency is high. Temperature monitoring points are set at key points in the system to monitor and control possible ignition points to ensure safe production. The hot water cleaning device can effectively prevent the occurrence of the Xu Ran phenomenon in the electrostatic tar collector. The pole matching method of the present invention adopts a mixed pole matching to improve the purification efficiency.

[0005] Obviously, the above-mentioned existing technologies focus on the establishment of environmental compliance monitoring systems and real-time monitoring of environmental pollution data. They are unable to effectively combine operation and maintenance records, emission concentration data and configuration parameter data to timely predict and determine whether changes in emission concentrations are related to operation and maintenance errors caused by parameter configuration deviations. Naturally, they are unable to take corresponding effective measures. What is particularly critical is that what is monitored is real-time environmental pollution data, and there is a lack of a detection mechanism for environmental pollution data in future time intervals, which makes it impossible for enterprises to further improve their environmental compliance. Summary of the invention

[0006] In order to solve the technical problems in the prior art, the present invention provides a big data analysis method for identifying compliance before and after operation and maintenance. Based on big data analysis, different mechanisms can be used to monitor the environmental pollution compliance of the emission gas in two different time intervals, namely, the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, and the shutdown time interval after the shutdown unit operation and maintenance operation of the thermal power unit of the target power plant before and after operation and maintenance. It is particularly critical that for the monitoring of environmental pollution compliance in the shutdown time interval after the shutdown unit operation and maintenance operation, a customized artificial intelligence model is used to intelligently analyze the time required for the dust concentration of the gas around the circular chimney of the thermal power unit of the target power plant to drop below the more stringent lower set concentration threshold required for the shutdown time interval after the shutdown unit operation and maintenance operation is executed, and the time required for the acid gas content of the gas around the circular chimney to drop below the more stringent lower set content threshold required for the shutdown time interval, and then decide whether to trigger the emission violation signal associated with the shutdown time interval, thereby realizing the overall and reliable identification of the environmental pollution compliance of the target power plant before and after operation and maintenance, and ensuring the environmental protection compliance of the target power plant at all times.

[0007] According to the present invention, a big data analysis method for identifying compliance before and after operation and maintenance is provided, and the method is applied to an environmental monitoring station, and the method comprises:

[0008] Measuring the dust concentration and acid gas content of the circular chimney exhaust gas of the thermal power unit of the target power plant during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, so as to trigger an emission violation signal or an emission compliance signal associated with the working time interval; obtaining various configuration information of the thermal power unit during the working time interval;

[0009] Using the first big data analysis node to perform multiple learning on the deep neural network within the working time interval to obtain a deep neural network after completing multiple learning and output it as an intelligent emission analysis model, wherein the number of times the deep neural network completes learning is positively correlated with the radius of the circular chimney;

[0010] After detecting that the thermal power units of the target power plant have been shut down for maintenance, the second big data analysis node uses an intelligent emission analysis model to intelligently analyze the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold after the thermal power units of the target power plant have been shut down for maintenance, based on the various configuration information of the thermal power units, the dust concentration and acid gas content of the exhaust gas from each circular chimney corresponding to each moment evenly spaced within the working time interval, and the length of the time interval between two adjacent moments, and output them as the first prediction duration and the second prediction duration, respectively.

[0011] Compared with the prior art, the present invention has at least the following six key inventive features:

[0012] First: Based on big data analysis, different mechanisms are used to monitor the environmental pollution compliance of the emission gas of the thermal power units of the target power plant during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, and the shutdown time interval after the shutdown unit operation and maintenance operation, i.e. before and after operation and maintenance, so as to achieve a holistic and reliable identification of the environmental pollution compliance of the target power plant before and after operation and maintenance, and ensure the environmental protection compliance of the target power plant at all times;

[0013] Second: For the monitoring of environmental pollution compliance during the outage time interval after the shutdown of the unit operation and maintenance operation, a customized artificial intelligence model is used to intelligently analyze the time required for the dust concentration of the gas around the circular chimney to drop below the more stringent lower set concentration threshold required for the outage time interval after the thermal power unit of the target power plant is shut down for operation and maintenance, and the time required for the acid gas content of the gas around the circular chimney to drop below the more stringent lower set content threshold required for the outage time interval, and output them as the first prediction time and the second prediction time, respectively, so as to trigger the emission violation signal associated with the outage time interval when the first prediction time is greater than or equal to the set concentration drop time or the second prediction time is greater than or equal to the set content drop time, otherwise, trigger the emission compliance signal associated with the outage time interval;

[0014] Third: using a big data analysis node to perform multiple learning on a deep neural network within a working time interval to obtain a deep neural network after multiple learning and output it as an intelligent emission analysis model. The number of times the deep neural network completes learning is positively correlated with the radius of the circular chimney, thereby completing the customized structural design of an artificial intelligence model for intelligent analysis;

[0015] Fourthly, in each learning of the deep neural network, the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold after a certain shutdown operation of the thermal power unit of the known target power plant in history and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold are used as two output contents of the deep neural network, and various configuration information configured when the thermal power unit was most recently started before the shutdown operation was performed at a certain time in history, the dust concentration and acid gas content of each circular chimney exhaust gas corresponding to each time interval evenly spaced in the latest working time interval before the shutdown operation was performed at a certain time in history, and the time interval length between two adjacent times in the latest working time interval are used as multiple input contents of the deep neural network, thereby ensuring the learning effect of each learning of the deep neural network;

[0016] Fifth: obtain the hourly power generation, rated power, rated voltage, hourly coal consumption, coal combustion temperature and circular chimney radius of the thermal power unit configured when starting the unit operation and maintenance operation within the working time interval as various configuration information of the thermal power unit, and at the same time obtain the dust concentration and acid gas content of each circular chimney exhaust gas corresponding to each time interval evenly spaced within the working time interval and the time interval between two adjacent times, so as to complete the targeted screening of multiple basic information for intelligent analysis;

[0017] The sixth part: for the monitoring of environmental pollution compliance during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, the dust concentration and acid gas content of the circular chimney exhaust gas of the thermal power units of the target power plant are measured during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, and when the measured dust concentration exceeds or is equal to the first set concentration threshold or the acid gas content exceeds or is equal to the first set content threshold, the emission violation signal associated with the working time interval is triggered, otherwise, the emission compliance signal associated with the working time interval is triggered. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein:

[0019] Figure 1 A technical flow chart of a big data analysis method for identifying compliance before and after operation and maintenance according to the present invention;

[0020] Figure 2 This is a schematic diagram of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Embodiment 1 of the present invention;

[0021] Figure 3 This is a schematic diagram of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Embodiment 2 of the present invention;

[0022] Figure 4 This is a schematic diagram of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Embodiment 3 of the present invention;

[0023] Figure 5 This is a schematic diagram of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Embodiment 4 of the present invention;

[0024] Figure 6 This is a schematic diagram of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Embodiment 5 of the present invention;

[0025] Figure 7 This is a schematic diagram of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Embodiment 6 of the present invention;

[0026] Figure 8 This is a flowchart of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Example 7 of the present invention. DETAILED DESCRIPTION

[0027] like Figure 1As shown, a technical flow chart of a big data analysis method for identifying compliance before and after operation and maintenance according to the present invention is given.

[0028] Figure 1 In the paper, a working diagram of the environmental monitoring station for monitoring the target power plant is given. Figure 1 The target power plant includes a circular chimney of a thermal power unit, and the implementation subject of a big data analysis method for identifying compliance before and after operation and maintenance shown in the present invention is an environmental monitoring station;

[0029] like Figure 1 As shown, the specific technical process of the present invention is as follows:

[0030] Technical process 1: Targeted monitoring and processing of the environmental pollution compliance of the emission gases of the thermal power units of the target power plant during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation;

[0031] Specifically, the dust concentration and acid gas content of the circular chimney exhaust gas of the thermal power unit of the target power plant are measured during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, and when the measured dust concentration exceeds or is equal to a first set concentration threshold or the acid gas content exceeds or is equal to the first set content threshold, an emission violation signal associated with the working time interval is triggered, otherwise, an emission compliance signal associated with the working time interval is triggered;

[0032] like Figure 1 As shown, the technical process 1 adopts the real-time measurement method of physical quantities, and takes the peak value of dust concentration and the peak value of acid gas content in the working time interval as the dust concentration and acid gas content of the circular chimney exhaust gas of the thermal power unit in the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation of the thermal power unit of the target power plant;

[0033] Technical process 2: Introduce a customized artificial intelligence model to monitor environmental pollution compliance during the downtime period after the shutdown of the unit operation and maintenance, and use it to intelligently analyze how long it will take for the gas pollution level around the thermal power unit to reach compliance after the shutdown of the unit operation and maintenance;

[0034] Specifically, the degree of gas pollution around the thermal power unit depends on the specific value of the dust concentration and the specific value of the acid gas content of the gas around the circular chimney of the thermal power unit of the target power plant;

[0035] For example, the artificial intelligence model is an intelligent emission analysis model, and the customization is reflected in the following aspects:

[0036] First aspect: the intelligent emission analysis model is a deep neural network that has completed multiple learnings;

[0037] Second aspect: the number of times the deep neural network completes learning is positively correlated with the radius of the circular chimney, thereby customizing artificial intelligence models with different structures for settings of different circular chimney radii;

[0038] The third aspect: in each learning performed on the deep neural network, the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold after a known shutdown operation of the thermal power unit of the target power plant in history and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold are used as two output contents of the deep neural network, and the various configuration information configured when the thermal power unit was most recently started before the shutdown operation was performed at a certain time in history, the dust concentration and acid gas content of each circular chimney exhaust gas corresponding to each time interval evenly spaced in the latest working time interval before the shutdown operation was performed at a certain time in history, and the time interval length between two adjacent times in the latest working time interval are used as multiple input contents of the deep neural network, thereby ensuring the learning effect of each learning performed on the deep neural network;

[0039] Technical process three: Targeted screening of multiple basic information for environmental pollution compliance during the downtime period after the shutdown of the unit operation and maintenance operations to ensure the reliability and stability of the intelligent analysis results;

[0040] Specifically, the multiple pieces of basic information are various configuration information of the thermal power unit, dust concentration and acid gas content of the exhaust gas of each circular chimney corresponding to each time at even intervals within the working time interval, and the time interval between two adjacent times;

[0041] More specifically, the various configuration information of the thermal power unit is the hourly power generation, rated power, rated voltage, hourly coal consumption, coal combustion temperature and circular chimney radius configured when the thermal power unit is started for operation and maintenance;

[0042] It can be seen that different configuration information can obtain different intelligent analysis results. In this way, when the monitoring result of environmental pollution compliance in the downtime interval after the shutdown operation of the unit in the subsequent intelligent analysis does not meet expectations, the monitoring result of environmental pollution compliance in the downtime interval after the shutdown operation of the unit can be intelligently analyzed again by adjusting the configuration information until it meets expectations, thereby timely predicting and judging whether the change in emission concentration is related to the operation and maintenance errors of parameter configuration deviation, realizing timely correction of the operation and maintenance errors of parameter configuration deviation, and ensuring environmental pollution compliance in the downtime interval after the shutdown operation of the unit;

[0043] Technical process 4: The intelligent emission analysis model with customized structure design in technical process 2 is used to intelligently analyze how long it will take for the gas pollution level around the thermal power unit to reach compliance after the unit operation and maintenance operation is shut down based on multiple basic information selected in a targeted manner in technical process 3;

[0044] Specifically, the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold after the thermal power unit of the target power plant is shut down for maintenance are intelligently analyzed, and the first prediction time and the second prediction time are output respectively. The second set concentration threshold and the second set content threshold are both stricter lower thresholds that need to be used for compliance with the outage time interval requirements.

[0045] Further specifically, when the first predicted time period is greater than or equal to the set concentration drop time period or the second predicted time period is greater than or equal to the set concentration drop time period, an emission violation signal associated with the outage time interval is triggered, otherwise, an emission compliance signal associated with the outage time interval is triggered;

[0046] In summary, technical process 1 realizes the targeted monitoring and processing of the environmental pollution compliance of the emission gas of the thermal power units of the target power plant during the working time period between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation. Technical processes 2, 3 and 4 realize the targeted monitoring and processing of the environmental pollution compliance of the gas around the circular chimney of the thermal power units of the target power plant after the shutdown unit operation and maintenance operation. Among them, technical processes 2 and 3 use different big data analysis nodes respectively;

[0047] Specifically, when different mechanisms are used to monitor the environmental pollution compliance of the emission gas of the thermal power units of the target power plant during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, and the shutdown time interval after the shutdown unit operation and maintenance operation, i.e., before and after the operation and maintenance, the monitoring position of the emission gas is the same;

[0048] In this way, targeted identification of the compliance of the thermal power units of the target power plant before and after operation and maintenance was completed based on big data analysis.

[0049] The key points of the present invention are: monitoring and processing of different monitoring mechanisms for environmental pollution compliance of gas around circular chimneys before and after the shutdown of unit operation and maintenance operations, structural customization of an artificial intelligence model for intelligently analyzing how long it takes for the level of gas pollution around a thermal power unit to reach compliance after the shutdown of unit operation and maintenance operations, and targeted screening of basic information, and intelligently analyzing the monitoring results of environmental pollution compliance during the downtime period after the shutdown of unit operation and maintenance operations by adjusting the configuration information until expectations are met, so as to achieve timely correction of operation and maintenance errors caused by parameter configuration deviations.

[0050] The present invention will be described in detail below in the form of various specific embodiments.

[0051] Example 1

[0052] Figure 2 The present invention is a flowchart of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Example 1 of the present invention.

[0053] like Figure 2 As shown, the big data analysis method for identifying compliance before and after operation and maintenance also includes the following steps:

[0054] Step 21: measuring the dust concentration and acid gas content of the circular chimney exhaust gas of the thermal power unit of the target power plant during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, and triggering an emission violation signal associated with the working time interval when the measured dust concentration exceeds or is equal to a first set concentration threshold or the acid gas content exceeds or is equal to the first set content threshold, otherwise, triggering an emission compliance signal associated with the working time interval;

[0055] For example, in the dust concentration and acid gas content of the gas discharged from the circular chimney of the thermal power unit, the acid gas is sulfur dioxide gas and / or hydrogen chloride gas;

[0056] Specifically, the advantage of a round chimney is that it is better than a square one in terms of pressure bearing capacity and combustion conditions. If the outer shape is designed to be square, the inside is also round. In terms of pressure bearing, those skilled in the art know that stress concentration will not occur, eliminating the right-angle concentrated force structure, and the force in the round structure is uniform in all places, which can better ensure safety. The same is true for combustion. There will be no "dead corners" and insufficient air at the angles, which affects combustion. Therefore, the chimney of the unit is generally made into a round shape;

[0057] Step 22: obtaining the hourly power generation, rated power, rated voltage, hourly coal consumption, coal combustion temperature and circular chimney radius of the thermal power unit configured when starting the unit operation and maintenance operation within the working time interval as various configuration information of the thermal power unit;

[0058] Specifically, the hourly power generation, rated power, rated voltage, hourly coal consumption, coal combustion temperature and circular chimney radius configured when the thermal power unit is started up for maintenance operation are obtained within the working time interval as various configuration information of the thermal power unit, including: the circular chimney radius of the thermal power unit is adjustable, so that it can be used as a changed configuration information to perform feedback and intelligent control of the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold after the thermal power unit of the target power plant is shut down for maintenance operation and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold;

[0059] Step 23: using the first big data analysis node to perform multiple learning on the deep neural network within the working time interval to obtain a deep neural network after completing multiple learning and output it as an intelligent emission analysis model, wherein the number of times the deep neural network completes learning is positively correlated with the radius of the circular chimney;

[0060] For example, the number of times the deep neural network completes learning is positively correlated with the radius of the circular chimney, including: when the radius of the circular chimney is 0.5 meters, the number of times the deep neural network completes learning is 100 times, when the radius of the circular chimney is 0.75 meters, the number of times the deep neural network completes learning is 150 times, when the radius of the circular chimney is 1.0 meters, the number of times the deep neural network completes learning is 200 times, and so on;

[0061] Generally, the circular chimney should be more than 5 meters higher than the buildings within a radius of 200 meters. If this requirement cannot be met, the emission rate should be increased by 50%. The chimney of a new pollution source should generally not be lower than 15 meters. If the chimney of a new pollution source is lower than 15 meters, the emission rate standard value should be extrapolated and then increased by 50%;

[0062] Step 24: after detecting that the thermal power unit of the target power plant has been shut down for maintenance, the second big data analysis node uses an intelligent emission analysis model to intelligently analyze the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold after the thermal power unit of the target power plant has been shut down for maintenance, based on various configuration information of the thermal power unit, the dust concentration and acid gas content of the exhaust gas of each circular chimney corresponding to each time interval evenly spaced within the working time interval, and the time interval between two adjacent times, and output them as the first prediction time and the second prediction time, respectively;

[0063] Specifically, when different mechanisms are used to monitor the environmental pollution compliance of the emission gas of the thermal power units of the target power plant during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, and the shutdown time interval after the shutdown unit operation and maintenance operation, i.e., before and after the operation and maintenance, the monitoring position of the emission gas is the same;

[0064] Step 25: When the first prediction time is greater than or equal to the set concentration drop time or the second prediction time is greater than or equal to the set concentration drop time, trigger an emission violation signal associated with the outage time interval; otherwise, trigger an emission compliance signal associated with the outage time interval;

[0065] Wherein, the first set concentration threshold is greater than the second set concentration threshold, and the first set content threshold is greater than the second set content threshold;

[0066] Specifically, the first set concentration threshold is greater than the second set concentration threshold, and the reason why the first set content threshold is greater than the second set content threshold is that the second set concentration threshold and the second set content threshold are for the time interval after the shutdown of the unit operation and maintenance operation, and the numerical requirements are stricter, so the thresholds should be lower. On the contrary, the first set concentration threshold and the first set content threshold are for the time interval after the startup of the unit operation and maintenance operation, and the numerical requirements are looser, so the thresholds should be higher.

[0067] Among them, an intelligent emission analysis model is used to intelligently analyze the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold after the thermal power unit of the target power plant is shut down for unit maintenance based on various configuration information of the thermal power unit, the dust concentration and acid gas content of each circular chimney emission gas corresponding to each moment evenly spaced within the working time interval, and the time interval between two adjacent moments, and output them as the first prediction time and the second prediction time respectively, including: the number of moments evenly spaced within the working time interval is positively correlated with the hourly power generation configured by the thermal power unit when the unit is started for unit maintenance;

[0068] For example, the number of moments evenly spaced apart from each other within the working time interval is positively correlated with the hourly power generation configured when the thermal power unit is started for operation and maintenance, including: the smaller the hourly power generation configured when the thermal power unit is started for operation and maintenance, the smaller the number of moments evenly spaced apart from each other within the working time interval, and accordingly, the longer the time interval between two adjacent moments when the working time interval remains unchanged;

[0069] And wherein, a first big data analysis node is used to perform multiple learning on the deep neural network within the working time interval to obtain a deep neural network after completing multiple learning and output it as an intelligent emission analysis model, and the number of times the deep neural network completes learning is positively correlated with the circular chimney radius, including: in each learning performed on the deep neural network, the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold after a known shutdown operation of the thermal power unit of the target power plant in history and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold are used as two output contents of the deep neural network, and various configuration information configured when the thermal power unit was most recently started before the shutdown operation was performed at a certain time in history, the dust concentration and acid gas content of each circular chimney emission gas corresponding to each time evenly spaced in the latest working time interval before the shutdown operation was performed at a certain time in history, and the length of the time interval between two adjacent times in the latest working time interval are used as multiple input contents of the deep neural network to complete this learning performed on the deep neural network.

[0070] Example 2

[0071] Figure 3 This is a schematic flowchart of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Example 2 of the present invention.

[0072] like Figure 3 As shown, compared with Figure 2 , when measuring the dust concentration and acid gas content of the circular chimney exhaust gas of the thermal power unit of the target power plant during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, and when the measured dust concentration exceeds or is equal to the first set concentration threshold or the acid gas content exceeds or is equal to the first set content threshold, triggering the emission violation signal associated with the working time interval, otherwise, before triggering the emission compliance signal associated with the working time interval, that is, before step 21, the big data analysis method for identifying compliance before and after operation and maintenance also includes the following steps:

[0073] Step 31: starting the operation recording device to record the operation records corresponding to each operation and maintenance operation of the thermal power unit of the target power plant;

[0074] For example, the startup operation recording device is used to record each operation record corresponding to each operation and maintenance operation of the thermal power unit of the target power plant, including: the operation recording device is a FLASH flash memory device.

[0075] Example 3

[0076] Figure 4 This is a schematic flow chart of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Example 3 of the present invention.

[0077] like Figure 4 As shown, compared with Figure 2 , when the first predicted time is greater than or equal to the set concentration drop time or the second predicted time is greater than or equal to the set content drop time, the emission violation signal associated with the outage time interval is triggered; otherwise, after the emission compliance signal associated with the outage time interval is triggered, that is, after step 25, the big data analysis method for identifying compliance before and after operation and maintenance further includes the following steps:

[0078] Step 41: when an emission violation signal associated with the outage time interval is received, triggering a configuration information adjustment operation on various configuration information of the thermal power unit to perform intelligent analysis again until no emission violation signal associated with the outage time interval is received;

[0079] Since different configuration information can obtain different intelligent analysis results, when the monitoring results of environmental pollution compliance in the downtime period after the subsequent intelligent analysis of the shutdown unit operation and maintenance operation do not meet expectations, the monitoring results of environmental pollution compliance in the downtime period after the shutdown unit operation and maintenance operation can be intelligently analyzed again by adjusting the configuration information until the expectations are met, thereby timely predicting and judging whether the change in emission concentration is related to the operation and maintenance errors of parameter configuration deviation, realizing timely correction of operation and maintenance errors of parameter configuration deviation, and ensuring environmental pollution compliance in the downtime period after the shutdown unit operation and maintenance operation.

[0080] Example 4

[0081] Figure 5 This is a schematic flow chart of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Example 4 of the present invention.

[0082] like Figure 5 As shown, compared with Figure 2 , when the first predicted time is greater than or equal to the set concentration drop time or the second predicted time is greater than or equal to the set content drop time, the emission violation signal associated with the outage time interval is triggered; otherwise, after the emission compliance signal associated with the outage time interval is triggered, that is, after step 25, the big data analysis method for identifying compliance before and after operation and maintenance further includes the following steps:

[0083] Step 51: wirelessly transmitting an emission violation signal associated with the outage time interval or an emission compliance signal associated with the outage time interval to a remote environmental monitoring server via a wireless communication line;

[0084] For example, wirelessly transmitting an emission violation signal associated with an outage time interval or an emission compliance signal associated with an outage time interval to a remote environmental monitoring server via a wireless communication link includes: the wireless communication link is a frequency division duplex communication link or a time division duplex communication link.

[0085] Example 5

[0086] Figure 6 This is a schematic flowchart of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Example 5 of the present invention.

[0087] like Figure 6 As shown, compared with Figure 2 , when the first predicted time is greater than or equal to the set concentration drop time or the second predicted time is greater than or equal to the set content drop time, the emission violation signal associated with the outage time interval is triggered; otherwise, after the emission compliance signal associated with the outage time interval is triggered, that is, after step 25, the big data analysis method for identifying compliance before and after operation and maintenance further includes the following steps:

[0088] Step 61: using a field display device to display in real time an emission violation signal associated with the outage time interval or an emission compliance signal associated with the outage time interval;

[0089] For example, using an on-site display device to display in real time an emission violation signal associated with an outage time interval or an emission compliance signal associated with an outage time interval includes: the on-site display device is an LED display array or an LCD display array.

[0090] Example 6

[0091] Figure 7 This is a schematic flowchart of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Example 6 of the present invention.

[0092] like Figure 7 As shown, compared with Figure 2 , when the first predicted time is greater than or equal to the set concentration drop time or the second predicted time is greater than or equal to the set content drop time, the emission violation signal associated with the outage time interval is triggered; otherwise, after the emission compliance signal associated with the outage time interval is triggered, that is, after step 25, the big data analysis method for identifying compliance before and after operation and maintenance further includes the following steps:

[0093] Step 71: using a voice alarm device to perform on-site broadcasting of a voice alarm file corresponding to the emission violation signal associated with the outage time interval when receiving the emission violation signal associated with the outage time interval;

[0094] Alternatively, an optical alarm device may be selected to flash the lamp body at a flashing frequency corresponding to the emission violation signal associated with the outage time interval when the emission violation signal associated with the outage time interval is received.

[0095] Example 7

[0096] Figure 8 This is a flowchart of the steps of a big data analysis method for identifying compliance before and after operation and maintenance according to Example 7 of the present invention.

[0097] like Figure 8 As shown, compared with Figure 2 , after the first big data analysis node is used to perform multiple learning on the deep neural network within the working time interval to obtain the deep neural network after completing multiple learning and output it as the intelligent emission analysis model, and the number of times the deep neural network completes learning is positively correlated with the radius of the circular chimney, that is, after step 23, the big data analysis method for identifying compliance before and after operation and maintenance also includes the following steps:

[0098] Step 81: using a model storage device to complete model storage of the intelligent emission analysis model by storing various model parameters of the intelligent emission analysis model;

[0099] For example, using a model storage device to complete model storage of the intelligent emission analysis model by storing various model parameters of the intelligent emission analysis model includes: using a TF storage device or an MMC storage device to complete model storage of the intelligent emission analysis model by storing various model parameters of the intelligent emission analysis model.

[0100] Next, various embodiments of the present invention will be described in detail.

[0101] In a big data analysis method for identifying compliance before and after operation and maintenance according to various embodiments of the present invention:

[0102] The first big data analysis node is used to perform multiple learning on the deep neural network within the working time interval to obtain the deep neural network after completing the multiple learning and output it as the intelligent emission analysis model, wherein the number of times the deep neural network completes the learning is positively correlated with the radius of the circular chimney, including: using a numerical conversion function to represent the numerical conversion relationship of the number of times the deep neural network completes the learning and the radius of the circular chimney;

[0103] Specifically, a programmable logic device may be selected to implement simulation and testing of the numerical conversion function;

[0104] And wherein, a numerical conversion relationship in which a numerical conversion function is used to represent the positive correlation between the number of times the deep neural network completes learning and the circular chimney radius includes: in the numerical conversion function, the circular chimney radius is the input value of the numerical conversion function, and the number of times the deep neural network completes learning corresponding to the circular chimney radius is the output value of the numerical conversion function.

[0105] And in a big data analysis method for identifying compliance before and after operation and maintenance according to various embodiments of the present invention:

[0106] The intelligent emission analysis model is used to intelligently analyze the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold after the thermal power unit of the target power plant is shut down for unit maintenance based on various configuration information of the thermal power unit, the dust concentration and acid gas content of the exhaust gas of each circular chimney corresponding to each time interval evenly spaced within the working time interval, and the time interval between two adjacent times, and output them as the first prediction time and the second prediction time, respectively. It also includes: performing numerical normalization processing on various configuration information of the thermal power unit, the dust concentration and acid gas content of the exhaust gas of each circular chimney corresponding to each time interval evenly spaced within the working time interval, and the time interval between two adjacent times, and then inputting them into the intelligent emission analysis model in parallel;

[0107] Among them, performing numerical normalization processing on various configuration information of the thermal power unit, dust concentration and acid gas content of each circular chimney exhaust gas corresponding to each time interval evenly spaced within the working time interval, and the time interval length between two adjacent times, and then inputting them in parallel into the intelligent emission analysis model includes: performing binary numerical conversion processing on various configuration information of the thermal power unit, dust concentration and acid gas content of each circular chimney exhaust gas corresponding to each time interval evenly spaced within the working time interval, and the time interval length between two adjacent times, and then inputting them in parallel into the intelligent emission analysis model;

[0108] Specifically, a CPLD device can be selected to perform binary value conversion processing on various configuration information of the thermal power unit, the dust concentration and acid gas content of each circular chimney exhaust gas corresponding to each time interval evenly spaced within the working time interval, and the time interval length between two adjacent times, and then input them into the intelligent emission analysis model in parallel;

[0109] Among them, an intelligent emission analysis model is used to intelligently analyze the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold after the thermal power unit of the target power plant is shut down for operation and maintenance based on various configuration information of the thermal power unit, the dust concentration and acid gas content of each circular chimney emission gas corresponding to each moment evenly spaced within the working time interval, and the time interval between two adjacent moments, and output them as the first prediction time and the second prediction time respectively, and also includes: the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold after the thermal power unit of the target power plant is shut down for operation and maintenance output by the intelligent emission analysis model are expressed in the form of normalized values;

[0110] And wherein, the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold after the thermal power units of the target power plant output by the intelligent emission analysis model are shut down for maintenance, and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold are numerically normalized representations, including: the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold after the thermal power units of the target power plant output by the intelligent emission analysis model are shut down for maintenance, and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold are representations after binary value conversion.

[0111] In addition, the present invention may also cite the following specific technical contents to highlight the outstanding essential features and significant technical progress of the present invention:

[0112] The method further comprises: using a first big data analysis node to perform multiple learning on a deep neural network within a working time interval to obtain a deep neural network after multiple learning and output it as an intelligent emission analysis model, wherein the number of times the deep neural network completes learning is positively correlated with the radius of the circular chimney, and further comprises: using a numerical simulation mode to implement multiple learning on the deep neural network within a working time interval to obtain a deep neural network after multiple learning and output it as a data processing process of the intelligent emission analysis model;

[0113] For example, the use of a numerical simulation mode to implement multiple learning of a deep neural network within a working time interval to obtain a deep neural network after multiple learning and test and simulate the data processing process as the output of the intelligent emission analysis model includes: using a MATLAB toolbox to implement multiple learning of a deep neural network within a working time interval to obtain a deep neural network after multiple learning and test and simulate the data processing process as the output of the intelligent emission analysis model;

[0114] Wherein, the dust concentration and acid gas content of the circular chimney exhaust gas of the thermal power unit of the target power plant are measured during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, and when the measured dust concentration exceeds or is equal to the first set concentration threshold or the acid gas content exceeds or is equal to the first set content threshold, the emission violation signal associated with the working time interval is triggered; otherwise, the emission compliance signal associated with the working time interval is triggered, including: taking the peak value of the dust concentration of the circular chimney exhaust gas of the thermal power unit during the working time interval as the measured dust concentration, and taking the peak value of the acid gas content of the circular chimney exhaust gas of the thermal power unit during the working time interval as the measured acid gas content;

[0115] And wherein, obtaining the hourly power generation, rated power, rated voltage, hourly coal consumption, coal combustion temperature and circular chimney radius of the thermal power unit configured when starting the unit operation and maintenance operation within the working time interval as various configuration information of the thermal power unit includes: using different physical quantity measurement elements to complete the synchronous measurement of the hourly power generation, coal combustion temperature and circular chimney radius respectively.

[0116] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A big data analysis method for identifying compliance before and after operation and maintenance, applied to environmental monitoring stations, characterized in that: The method comprises: Measuring the dust concentration and acid gas content of the circular chimney exhaust gas of the thermal power units of the target power plant during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, so as to trigger an emission violation signal or an emission compliance signal associated with the working time interval; Acquire various configuration information of the thermal power unit during the working time interval, wherein the various configuration information of the thermal power unit is the hourly power generation, rated power, rated voltage, hourly coal consumption, coal combustion temperature and circular chimney radius configured when the thermal power unit is started for operation and maintenance; Using the first big data analysis node to perform multiple learning on the deep neural network within the working time interval to obtain a deep neural network after completing multiple learning and output it as an intelligent emission analysis model, wherein the number of times the deep neural network completes learning is positively correlated with the radius of the circular chimney; After detecting that the thermal power unit of the target power plant has been shut down for maintenance, the second big data analysis node uses an intelligent emission analysis model to intelligently analyze the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold after the thermal power unit of the target power plant has been shut down for maintenance, based on various configuration information of the thermal power unit, the dust concentration and acid gas content of the exhaust gas of each circular chimney corresponding to each time interval evenly spaced within the working time interval, and the time interval between two adjacent times, and output them as the first prediction time and the second prediction time, respectively; Wherein, in each learning performed on the deep neural network, the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold after a certain shutdown operation of the thermal power unit of the known target power plant in history and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold are used as two output contents of the deep neural network, and various configuration information configured when the thermal power unit was most recently started before the shutdown operation was performed in the history, the dust concentration and acid gas content of each circular chimney exhaust gas corresponding to each time evenly spaced in the latest working time interval before the shutdown operation was performed in the history, and the time interval length between two adjacent times in the latest working time interval are used as multiple input contents of the deep neural network to complete the current learning performed on the deep neural network; Among them, the number of moments evenly spaced in the working time interval is positively correlated with the hourly power generation configured when the thermal power unit is started for unit operation and maintenance operations; Wherein, when the measured dust concentration exceeds or is equal to a first set concentration threshold or the acid gas content exceeds or is equal to a first set content threshold, an emission violation signal associated with the working time interval is triggered; otherwise, an emission compliance signal associated with the working time interval is triggered; Wherein, when the first prediction time is greater than or equal to the set concentration drop time or the second prediction time is greater than or equal to the set content drop time, the emission violation signal associated with the outage time interval is triggered, otherwise, the emission compliance signal associated with the outage time interval is triggered; Wherein, the first set concentration threshold is greater than the second set concentration threshold, and the first set content threshold is greater than the second set content threshold; Among them, a numerical conversion function is used to represent the numerical conversion relationship between the number of times the deep neural network completes learning and the positive correlation between the circular chimney radius, the circular chimney radius is the input value of the numerical conversion function, and the number of times the deep neural network completes learning corresponding to the circular chimney radius is the output value of the numerical conversion function.

2. A big data analysis method for identifying compliance before and after operation and maintenance as claimed in claim 1, characterized in that: The method further comprises: measuring the dust concentration and acid gas content of the circular chimney exhaust gas of the thermal power unit of the target power plant during the working time interval between the start-up unit operation and maintenance operation and the shutdown unit operation and maintenance operation, and triggering an emission violation signal associated with the working time interval when the measured dust concentration exceeds or is equal to a first set concentration threshold or the acid gas content exceeds or is equal to the first set content threshold; otherwise, before triggering the emission compliance signal associated with the working time interval, the method further comprises: The startup operation recording device is used to record the operation records corresponding to each operation and maintenance operation of the thermal power unit of the target power plant.

3. A big data analysis method for identifying compliance before and after operation and maintenance as claimed in claim 1, characterized in that: When the first predicted time period is greater than or equal to the set concentration drop time period or the second predicted time period is greater than or equal to the set concentration drop time period, an emission violation signal associated with the outage time interval is triggered; otherwise, after the emission compliance signal associated with the outage time interval is triggered, the method further includes: When an emission violation signal associated with the outage time interval is received, a configuration information adjustment operation on various configuration information of the thermal power unit is triggered to perform intelligent analysis again until an emission violation signal associated with the outage time interval is no longer received.

4. A big data analysis method for identifying compliance before and after operation and maintenance as claimed in claim 1, characterized in that: When the first predicted time period is greater than or equal to the set concentration drop time period or the second predicted time period is greater than or equal to the set concentration drop time period, an emission violation signal associated with the outage time interval is triggered; otherwise, after the emission compliance signal associated with the outage time interval is triggered, the method further includes: The emission violation signal associated with the outage time interval or the emission compliance signal associated with the outage time interval is wirelessly transmitted to a remote environment monitoring server through a wireless communication line.

5. A big data analysis method for identifying compliance before and after operation and maintenance as claimed in claim 1, characterized in that: When the first predicted time period is greater than or equal to the set concentration drop time period or the second predicted time period is greater than or equal to the set concentration drop time period, an emission violation signal associated with the outage time interval is triggered; otherwise, after the emission compliance signal associated with the outage time interval is triggered, the method further includes: An on-site display device is used to display in real time an emission violation signal associated with the outage time interval or an emission compliance signal associated with the outage time interval.

6. A big data analysis method for identifying compliance before and after operation and maintenance as claimed in claim 1, characterized in that: When the first predicted time period is greater than or equal to the set concentration drop time period or the second predicted time period is greater than or equal to the set concentration drop time period, an emission violation signal associated with the outage time interval is triggered; otherwise, after the emission compliance signal associated with the outage time interval is triggered, the method further includes: The voice alarm device is used to perform on-site broadcasting of a voice alarm file corresponding to the emission violation signal associated with the outage time interval when the emission violation signal associated with the outage time interval is received.

7. A big data analysis method for identifying compliance before and after operation and maintenance as claimed in claim 1, characterized in that: After the first big data analysis node is used to perform multiple learning on the deep neural network within the working time interval to obtain the deep neural network after completing the multiple learning and output it as the intelligent emission analysis model, and the number of times the deep neural network completes the learning is positively correlated with the radius of the circular chimney, the method further includes: The model storage device is used to store various model parameters of the intelligent emission analysis model to complete the model storage of the intelligent emission analysis model.

8. A big data analysis method for identifying compliance before and after operation and maintenance as claimed in any one of claims 1 to 7, characterized in that: The configuration information of the thermal power unit, the dust concentration and acid gas content of the exhaust gas of each circular chimney corresponding to each time interval evenly spaced within the working time interval, and the time interval between two adjacent times are numerically normalized and then input into the intelligent emission analysis model in parallel; Among them, various configuration information of the thermal power unit, the dust concentration and acid gas content of each circular chimney exhaust gas corresponding to each uniformly spaced moment within the working time interval, and the time interval between two adjacent moments are respectively converted into binary values ​​and then input into the intelligent emission analysis model in parallel; The time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold after the thermal power unit of the target power plant is shut down for maintenance and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold output by the intelligent emission analysis model are in the form of numerical normalization; Among them, the output of the intelligent emission analysis model shows that the time required for the dust concentration of the gas around the circular chimney to drop below the second set concentration threshold after the thermal power unit of the target power plant is shut down for maintenance, and the time required for the acid gas content of the gas around the circular chimney to drop below the second set content threshold are expressed in binary numerical conversion form.

Images