Nitrogen oxide emission prediction and ammonia injection control system and method based on fusion type artificial intelligence

Abstract

The invention discloses a system and a method for nitrogen oxide emission prediction and ammonia spraying control based on fusion type artificial intelligence, a plurality of electromagnetic valves are mounted at the lower end of an ammonia water tank, the other ends of the plurality of electromagnetic valves are respectively connected with one end of a spraying pipe, and a plurality of nozzles are mounted at the other end of the spraying pipe; the voltage-stabilized source, the data transmitting circuit and the control circuit are installed in the electric cabinet, the nitrogen oxide detector is installed in a monitoring area, and the nitrogen oxide detector detects nitrogen oxide data on site in real time and outputs different voltage signals to the signal input ends of the multi-path control circuit and the data transmitting circuit in real time according to the concentration of nitrogen oxide; the multi-path control circuit is electrically connected with the power input ends of the multiple electromagnetic valves so as to be connected with the electromagnetic valves of different numbers to be powered on, and then the flow of the ammonia water sprayed out of the spray head is adjusted. According to the specific magnitude of nitric oxide in the field environment, different electromagnetic valves are opened respectively, the purpose of effectively treating the field nitric oxide can be achieved, and energy is saved.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental protection dust reduction equipment, and in particular to a fusion-based artificial intelligence nitrogen oxide emission prediction and ammonia injection control system and method. Background Art

[0002] One of the main pollutants emitted by coal-fired power plants during operation is nitrogen oxides (NO). x In order to reduce its adverse impact on the environment, the existing technology generally uses an automated system to spray ammonia to neutralize and reduce the nitrogen oxide content in the exhaust gas. The currently used automatic ammonia spraying system generally includes multiple nozzles, ammonia storage tanks, solenoid valves and nitrogen oxide probes. When the probe detects nitrogen oxides in the monitored area, it controls the solenoid valve to open, and then pressurized ammonia water is sprayed out from multiple nozzles in a mist form to neutralize and dilute the nitrogen oxides on site.

[0003] Although the existing automatic ammonia spraying system can automatically spray ammonia based on the monitored nitrogen oxides, the opening and closing of its solenoid valve is fixed, that is, as long as the probe detects the presence of nitrogen oxides on site, regardless of the concentration of nitrogen oxides, the amount of ammonia sprayed is constant. When the amount of nitrogen oxides on site is small, the relatively large amount of ammonia sprayed will undoubtedly cause unnecessary waste. On the contrary, if the amount of nitrogen oxides on site is large and the amount of ammonia sprayed is relatively small, the nitrogen oxide removal effect will be poor, causing pollution to the environment. Therefore, there is an urgent need for a system that can automatically control the amount of ammonia sprayed based on the detected amount of nitrogen oxides on site, and can predict nitrogen oxide emissions based on the collected historical data. Summary of the invention

[0004] In order to overcome the shortcomings of the existing automatic ammonia spraying system, which is unable to accurately spray ammonia according to the amount of nitrogen oxides on site due to structural limitations, and has the disadvantages of wasting ammonia water and poor ammonia spraying effect, as well as being unable to predict nitrogen oxide emissions, the present invention provides a system that, under the joint action of relevant mechanisms and circuits, can open different numbers of solenoid valves according to the specific magnitude of nitrogen oxides in the on-site environment, so that ammonia water of different flow rates can be sprayed out through the nozzles respectively. The magnitude of the mist ammonia water sprayed is large when the nitrogen oxide concentration is large, and vice versa. Energy saving can be achieved while effectively treating the nitrogen oxides on site. The on-site data can also be transmitted remotely via a wireless mobile network in real time, so that remote managers can grasp the on-site nitrogen oxide data in real time. The relevant application units can also generate prediction data based on historical data through artificial intelligence learning, which provides a nitrogen oxide emission prediction and ammonia spraying control system and method based on fusion artificial intelligence, which provides favorable technical support for the formulation of relevant measures.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] The present invention provides a nitrogen oxide emission prediction and ammonia spraying control system based on fusion artificial intelligence, comprising a solenoid valve, an ammonia water tank, a spray tube, a voltage-stabilized power supply, a nitrogen oxide detector, a data transmission circuit, a control circuit, a data acquisition unit, a data analysis unit, a database unit, a data display unit, a data alarm unit and a prediction unit; a plurality of the solenoid valves are installed at the lower end of the ammonia water tank, the other ends of the plurality of the solenoid valves are respectively connected to one end of the spray tube, a plurality of nozzles are installed at the other end of the spray tube, the voltage-stabilized power supply, the data transmission circuit and the control circuit are installed in an electric control box, the nitrogen oxide detector is installed in a monitoring area, the nitrogen oxide detector detects the nitrogen oxide data on site in real time and outputs different voltage signals in real time according to the concentration of the nitrogen oxides to enter the multiple control circuits and the data transmission circuits. signal input end, the multiple control circuits are electrically connected to the power input ends of the multiple solenoid valves, respectively, so as to respectively connect different numbers of the solenoid valves to obtain electricity, thereby adjusting the flow rate of ammonia water sprayed from the nozzle; the data acquisition unit receives and stores the data sent by the data sending circuit, the data analysis unit analyzes the data based on the fusion artificial intelligence algorithm, the database unit stores the historical data of the emission time of nitrogen oxides at the production site and the emission amount at the corresponding time, the data display unit displays the specific nitrogen oxide emission data at the site on the display screen in real time, the data alarm unit can generate an alarm signal to prompt the on-site personnel when the nitrogen oxides exceed the standard at the site, and the prediction unit can use the pre-trained AI model according to the historical data stored in the database unit to issue a possible excessive nitrogen oxides warning in advance in the corresponding time period.

[0007] Furthermore, each control circuit includes an electrically connected resistor RN, an NPN transistor and a relay, the positive power input terminal of the relay is connected to the control power input terminal, the collector of the NPN transistor is connected to the negative power input terminal of the relay, and one end of the resistor RN is connected to the base of the NPN transistor.

[0008] Furthermore, the data sending circuit includes an electrically connected single-chip microcomputer module, a GPRS module and a resistor R1, the power input end of the single-chip microcomputer module is connected to the power input end of the GPRS module, the signal output end of the single-chip microcomputer module is connected to the signal input end of the GPRS module, and one end of the resistor R1 is electrically connected to the signal input end of the single-chip microcomputer module.

[0009] Furthermore, the power input terminals 1 and 2 of the voltage-stabilized power supply are electrically connected to the two poles of the AC 220V power supply, respectively; the power output terminals 3 and 4 of the voltage-stabilized power supply are electrically connected to the positive power input terminal of the relay, the emitter of the NPN transistor, the 1 and 2 pins of the single-chip microcomputer module, and the power input terminals 1 and 2 of the nitrogen oxide detector, respectively; the signal output terminal 3 of the nitrogen oxide detector is electrically connected to the other end of the resistor R1 and the other end of the resistor RN, respectively; the normally open contact terminal of the relay and the emitter of the NPN transistor are electrically connected to the power input terminals of the multiple solenoid valves, respectively.

[0010] Furthermore, five solenoid valves are provided, and five control circuits are provided. The five control circuits cooperatively control the five solenoid valves to work and have the same working principle.

[0011] Furthermore, the resistance value of the resistor R is 4K, and the resistance values ​​of the resistor RN of the first, second, third, fourth and fifth control circuits are 1.1K, 1.4M, 1.7M, 2M and 2.3M respectively.

[0012] Furthermore, the data alarm unit can push text messages to mobile phones of remote personnel when the nitrogen oxides on site exceed the standard.

[0013] Furthermore, the prediction unit selects 127 valid data as input after dimensionality reduction processing based on the multi-dimensional historical time series collected by the database unit to build a multi-data input fusion short-term prediction model that combines the traditional machine learning KNN algorithm and the deep learning neural network. The model can accurately predict the concentration of nitrogen oxides in different prediction time periods and realize quantitative control of the ammonia injection flow rate based on this data.

[0014] The present invention also provides a method for predicting nitrogen oxide emissions and controlling ammonia injection based on fused artificial intelligence. The method adopts the control system for predicting nitrogen oxide emissions and controlling ammonia injection based on fused artificial intelligence, and includes the following steps: after the AC power supply enters the power input end of the voltage-stabilized power supply, the voltage-stabilized power supply will output stable DC power from pins 3 and 4 under the action of the internal circuit, and enter the power input end of the multi-channel control circuit, the data sending circuit and the nitrogen oxide detector, and each circuit and the detector will be powered on and work; after the nitrogen oxide detector is powered on and work, pin 3 of the signal output end will output analog voltage signals of different sizes according to the different nitrogen oxide concentrations on site, and enter the other end of the resistor R1 and the other end of the resistor RN. After the single-chip microcomputer module inputs the dynamically changing analog voltage signal, it will convert the analog voltage signal into a digital signal, and then send it out through the GPRS module via the wireless mobile network. After the mobile phone or PC of the relevant personnel who has established a remote connection with the GPRS module receives the data, the display interface can intuitively display the nitrogen oxide concentration data on site.

[0015] Furthermore, the five solenoid valves are controlled by the five-way control circuit. The nitrogen oxide detector outputs a dynamically changing analog voltage signal according to the nitrogen oxide concentration at the scene. When the nitrogen oxide concentration data at the scene is very small, the voltage signal enters the base voltage of the NPN transistor after the voltage reduction and current limiting of the resistor RN of the first control circuit, and the voltage entering the resistor RN of the second to fifth control circuits is greater than 0.7V, and the voltage of the resistor RN is lower than 0.7V. At this time, the collector of the NPN transistor is turned on and outputs a low level to the negative power input terminal of the relay. The relay is energized to close its control power input terminal and the normally open contact terminal, and then the first solenoid valve is energized to open. Due to the small water flow, the amount of mist ammonia water sprayed by multiple nozzles is relatively small.

[0016] When the nitrogen oxide concentration data on site is small, the voltage signal enters the NPN transistor base voltage greater than 0.7V after the voltage reduction and current limiting of the resistor RN of the first control circuit and the second control circuit, and the voltage of the resistor RN entering the third to fifth control circuits is lower than 0.7V. At this time, the collector of the NPN transistor is turned on and the output of the low level enters the negative power input terminal of the relay. The relay is energized to close its control power input terminal and the normally open contact terminal, and then the first solenoid valve and the second solenoid valve are energized to open. Due to the slightly larger water flow, the amount of mist ammonia water sprayed by multiple nozzles becomes relatively larger;

[0017] When the nitrogen oxide concentration data on site is relatively large, the voltage signal enters the base voltage of the NPN transistor after the voltage reduction and current limiting of the resistor RN of the first, second and third control circuits, and the voltage of the resistor RN entering the fourth and fifth control circuits is greater than 0.7V, and the voltage of the resistor RN entering the fourth and fifth control circuits is lower than 0.7V. At this time, the collector of the NPN transistor is turned on and the output of the low level enters the negative power input terminal of the relay. The relay is energized to close its control power input terminal and the normally open contact terminal, and then the first, second and third solenoid valves are energized to open. Due to the large water flow, the amount of mist ammonia water sprayed by multiple nozzles becomes relatively large;

[0018] When the nitrogen oxide concentration data on site is very large, the voltage signal enters the NPN transistor base voltage greater than 0.7V after the voltage reduction and current limiting of the resistor RN of the first, second, third and fourth control circuits, and the voltage of the resistor RN entering the fifth control circuit is lower than 0.7V. At this time, the collector of the NPN transistor is turned on and the output of the low level enters the negative power input terminal of the relay. The relay is energized to close its control power input terminal and the normally open contact terminal, and then the first, second, third and fourth solenoid valves are energized to open. Due to the large water flow, the amount of mist ammonia water sprayed by multiple nozzles becomes relatively more.

[0019] When the on-site nitrogen oxide concentration data is the largest, the voltage signal enters the NPN transistor base voltage greater than 0.7V after being reduced and limited by the resistor RN of the first, second, third, fourth and fifth control circuits. At this time, the NPN transistor is turned on and the collector outputs a low level and enters the negative power input terminal of the relay. The relay is energized and its control power input terminal and normally open contact terminal are closed, and then the first, second, third, fourth and fifth solenoid valves are all energized and opened. Since the water flow is the largest, the amount of mist ammonia water sprayed from multiple nozzles is also at the maximum.

[0020] Compared with the prior art, the beneficial technical effects of the present invention are:

[0021] In the integrated artificial intelligence-based nitrogen oxide emission prediction and ammonia spraying control system and method of the present invention, the nitrogen oxide probe can detect the nitrogen oxide data on site (such as near the boiler of a coal-fired power plant) in real time, and output different voltage signals in real time according to the concentration of nitrogen oxides to enter the signal input end of the multi-channel control circuit and the data sending circuit. The multi-channel control circuit can respectively connect different numbers of solenoid valves to obtain electricity, and then ammonia water of different flow rates will be sprayed out through multiple nozzles. When the concentration of nitrogen oxides is high, the spray volume is large, and vice versa. The purpose of energy saving can be achieved while effectively treating the nitrogen oxides on site.

[0022] The data transmission circuit of the present invention can transmit the data on site via a wireless mobile network in real time, and the remote management personnel can grasp the nitrogen oxide data on site in real time. The relevant application unit can also generate prediction data based on historical data through artificial intelligence learning, which provides favorable technical support for the formulation of relevant measures (for example, according to historical data, it can be known that a large amount of nitrogen oxides will be discharged from a coal-fired power plant in a specific time period after one hour, so it can be checked in advance whether the ammonia water in the ammonia water tank is sufficient, or whether the working system is intact, to ensure that the equipment sprays ammonia normally to eliminate pollution. Based on the above settings, the present invention has good application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] 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 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.

[0024] Figure 1 It is a schematic diagram of the structure of the fusion-based artificial intelligence nitrogen oxide emission prediction and ammonia injection control system of the present invention;

[0025] Figure 2 A circuit diagram of the fusion-based artificial intelligence nitrogen oxide emission prediction and ammonia injection control system of the present invention;

[0026] Figure 3 This is a block diagram of the application software architecture of the present invention based on the fusion-type artificial intelligence nitrogen oxide emission prediction and ammonia injection control system.

[0027] Explanation of the accompanying drawings: 1, ammonia tank; 101, connecting pipe; 2, spray pipe; 3, nozzle; 4, data sending circuit; 5, control circuit; DC, solenoid valve; A1, voltage-stabilized power supply; A2, nitrogen oxide detector; A3, single-chip microcomputer module; A4, GPRS module; R1, resistor; RN, resistor; QN, NPN transistor; KN, relay. DETAILED DESCRIPTION

[0028] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0029] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically limited.

[0030] The technical solution provided by the embodiments of the present invention is described in detail below with reference to the accompanying drawings.

[0031] Example 1

[0032] like Figures 1 to 3 As shown, the fused artificial intelligence nitrogen oxide emission prediction and ammonia injection control system of embodiment 1 of the present invention includes a solenoid valve DC, an ammonia water tank 1 located at a high place (or pumped out by a pump), a spray pipe 2, a nozzle 3, a voltage-stabilized power supply A1 and a nitrogen oxide detector A2, and also has a data sending circuit 4, a control circuit 5, a data acquisition unit, a data analysis unit, a database unit, a data display unit, a data alarm unit and a prediction unit.

[0033] Specifically, Figure 1As shown, five solenoid valves DC are provided, nine nozzles 3 are provided, a connecting pipe 101 communicating with the interior of the ammonia tank 1 is installed at the lower end thereof, the lower end of the connecting pipe 101 is welded to the upper end of a horizontal pipe, one end of the five solenoid valves DC is laterally spaced a certain distance apart and is threadedly connected to the five branch pipes at the lower end of the horizontal pipe, the other end of the five solenoid valves DC is threadedly connected to the five upper pipes at the upper end of the spray pipe 2, respectively, the liquid inlet ends of the nine nozzles 3 are laterally spaced a certain distance apart and are threadedly connected to the nine lower pipes at the lower end of the spray pipe 2; a nitrogen oxide detector A2 is installed in the monitoring area, a voltage-stabilized power supply A1, a control circuit 5, and a data sending circuit 4 are installed on a circuit board in an electric control box 6, and a data acquisition unit, a data analysis unit, a database unit, a data display unit, a data alarm unit, and a prediction unit are application software installed in a PC, etc.

[0034] The working principle of the five control circuits in Embodiment 1 of the present invention to coordinately control the five solenoid valves is completely consistent. The following content will be representatively described by taking the working principle of one control circuit controlling one solenoid valve. The data acquisition unit can receive and store the data sent by the data sending circuit 4, the data analysis unit can analyze the data based on the fusion artificial intelligence algorithm, the database unit stores the historical data of the emission time of nitrogen oxides at the production site and the emission amount at the corresponding time, the data display unit can display the specific nitrogen oxide emission data at the site in real time through the display screen, the data alarm unit can generate an alarm signal to remind the site personnel when the nitrogen oxides at the site exceed the standard, and the prediction unit can use the pre-trained AI model based on the historical data stored in the database unit to issue a warning that the nitrogen oxides may exceed the standard in advance in the corresponding time period.

[0035] Preferably, the solenoid valve DC is a normally closed valve core solenoid valve with a power of 2W; the voltage-stabilized power supply A1 is a finished product of an AC 220V to DC 12V switching power supply module with a model of 220V / 12V; the nitrogen oxide detector A2 has two power input terminals and a signal output terminal. The signal output terminal will output an analog voltage signal varying between 0-5V depending on the detected on-site nitrogen oxide concentration.

[0036] And, if Figure 2 As shown, the data sending circuit 4 includes a single-chip microcomputer module A3, a GPRS module A4 and a resistor R1 connected via circuit board wiring, the power input terminal of the single-chip microcomputer module A3 is respectively connected to the power input terminal pins 1 and 2 of the GPRS module A4, the signal output terminal of the single-chip microcomputer module A3 is connected to the signal input terminal of the GPRS module A4, and one end of the resistor R1 is electrically connected to the signal input terminal pin 3 of the single-chip microcomputer module A3.

[0037] At the same time, if Figure 2As shown, each control circuit 5 includes a resistor RN, an NPN transistor QN, and a relay KN connected by circuit board wiring, the positive power input terminal of the relay KN is connected to the control power input terminal, the collector of the NPN transistor QN is connected to the negative power input terminal of the relay KN, and one end of the resistor RN is connected to the base of the NPN transistor QN. The resistance values ​​of the multi-channel control circuits are different. The data alarm unit can also push text messages to the mobile phones of remote personnel when the nitrogen oxides on site exceed the standard. The prediction unit is based on the historical time series of more than 600 dimensions of coal-fired power plants collected by the database unit, and 127 valid data selected after dimensionality reduction processing are used as input to build a multi-data input fusion short-term prediction model combining traditional machine learning KNN algorithm and deep learning neural network. It can accurately predict the concentration of nitrogen oxides in different prediction time periods and can also realize quantitative control of ammonia injection flow based on this data.

[0038] like Figure 2 As shown, the power input terminals 1 and 2 of the voltage stabilizer A1 are electrically connected to the two poles of the AC 220V power supply, and the power output terminals 3 and 4 of the voltage stabilizer A1 are electrically connected to the positive power input terminal of the power input terminal relay KN of the five-way control circuit and the emitter of the NPN transistor QN, the power input terminal of the data transmission circuit 4, the power input terminal single-chip module A3, and the power input terminal 1 and 2 of the nitrogen oxide detector A2. The signal output terminal 3 of the nitrogen oxide detector A2 is electrically connected to the other end of the signal input terminal resistor R1 of the data transmission circuit 4 and the other end of the signal input terminal resistor RN of the five-way control circuit, and the power output terminal relay KN of the five-way control circuit is electrically connected to the power input terminal of the five solenoid valves DC.

[0039] like Figure 2 As shown, after the 220V AC power enters the power input terminal of the voltage stabilizer A1, the voltage stabilizer A1 will output a stable DC 12V power supply from its 3rd and 4th pins under the action of its internal circuit to enter the five-way control circuit and data transmission circuit 4, and the power input terminal of the nitrogen oxide detector A2, so that the above circuits and detectors are powered on. After the nitrogen oxide detector A2 is powered on, its signal output terminal 3rd pin will output different analog voltage signals according to the different concentrations of nitrogen oxides on site, and enter the other end of the resistor R1 and the other end of the resistor RN of the five-way control circuit.

[0040] Through the existing mature Internet of Things technology, the single-chip computer module A3 will convert the analog voltage signal into a digital signal after inputting the dynamically changing analog voltage signal, and then send it out through the wireless mobile network via the GPRS module. After the mobile phone or PC of the relevant personnel who has established a connection with the GPRS module A4 at the remote end receives the data, the display interface can intuitively display the nitrogen oxide concentration data on site. When the concentration is high, the display interface displays large numbers or the waveform peak value is high, and when the concentration is low, the display interface displays small numbers or the waveform peak value is low. In this way, the remote monitoring personnel can understand the on-site data in real time.

[0041] In this embodiment 1, in the five-way control circuit, after the nitrogen oxide detector A2 outputs a dynamically changing analog voltage signal according to the nitrogen oxide concentration at the site at pin 3, when the nitrogen oxide concentration data at the site is very small, the voltage signal enters the base voltage of the NPN transistor QN after the resistor RN of the first control circuit is reduced and limited, and is greater than 0.7V (the voltage entering the resistor RN of the second to fifth control circuits is lower than 0.7V), so the collector of the NPN transistor QN is turned on and outputs a low level to the negative power input terminal of the relay KN, and the relay KN is energized to close its control power input terminal and the normally open contact terminal, and then the first solenoid valve DC is energized to open, and due to the small water flow, the amount of mist ammonia water sprayed from the nine nozzles is relatively small.

[0042] When the nitrogen oxide concentration data on site is small, the voltage signal enters the NPN transistor QN base voltage greater than 0.7V after being stepped down and limited by the resistor RN of the first and second control circuits (the voltage entering the resistor RN of the third to fifth control circuits is lower than 0.7V), and thus, the collector of the NPN transistor QN is turned on and outputs a low level which enters the negative power input terminal of the relay KN. The relay KN is energized to close its control power input terminal and the normally open contact terminal, and then the first and second solenoid valves DC are energized to open. As the water flow is slightly larger, the amount of mist ammonia water sprayed from the nine nozzles becomes relatively larger.

[0043] When the on-site nitrogen oxide concentration data is large, the voltage signal enters the NPN transistor QN base voltage greater than 0.7V after being reduced and limited by the resistor RN of the first, second and third control circuits (the voltage entering the resistor RN of the fourth and fifth control circuits is lower than 0.7V), so the NPN transistor QN is turned on, the collector outputs a low level and enters the negative power input terminal of the relay KN, and the relay KN is energized to close its control power input terminal and the normally open contact terminal, and then the first, second and third solenoid valves DC are energized to open. Due to the large water flow, the amount of mist ammonia water sprayed by the nine nozzles becomes relatively large.

[0044] When the on-site nitrogen oxide concentration data is very large, the voltage signal enters the NPN transistor QN base voltage greater than 0.7V after being reduced and limited by the resistor RN of the first, second, third and fourth control circuits (the voltage entering the resistor RN of the fifth control circuit is lower than 0.7V), so the NPN transistor QN is turned on and the collector outputs a low level and enters the negative power input terminal of the relay KN. The relay KN is energized and its control power input terminal and the normally open contact terminal are closed, and then the first, second, third and fourth solenoid valves DC are energized and opened. Due to the large water flow, the amount of mist ammonia water sprayed by the nine nozzles becomes relatively more.

[0045] When the on-site nitrogen oxide concentration data is the largest, the voltage signal enters the NPN transistor QN base voltage greater than 0.7V after being reduced and limited by the resistor RN of the first, second, third, fourth and fifth control circuits. Therefore, the collector of the NPN transistor QN is turned on and outputs a low level which enters the negative power input terminal of the relay KN. The relay KN is energized to close its control power input terminal and the normally open contact terminal, and then the first, second, third, fourth and fifth solenoid valves DC are all energized to open. Since the water flow is the largest, the amount of mist ammonia water sprayed from the nine nozzles is also at the maximum.

[0046] At this time, the nitrogen oxide probe in Example 1 of the present invention can detect the nitrogen oxide data on site (such as near the boiler of a coal-fired power plant) in real time, and output different voltage signals in real time according to the concentration of nitrogen oxides to enter the signal input end of the multi-channel control circuit and the data sending circuit. The multi-channel control circuit can respectively connect different numbers of solenoid valves to obtain electricity, and then ammonia water with different flow rates will be sprayed out through multiple nozzles. When the concentration of nitrogen oxides is high, the spray volume is large, and vice versa. The purpose of energy saving can be achieved while effectively treating the nitrogen oxides on site.

[0047] In this embodiment 1, the data transmission circuit can transmit the on-site data remotely via the wireless mobile network in real time, and the remote management personnel can grasp the on-site nitrogen oxide data in real time. The relevant application unit can also generate prediction data based on historical data through artificial intelligence learning, which plays a favorable technical support for the formulation of relevant measures (for example, according to historical data, it can be known that a large amount of nitrogen oxides will be discharged from the coal-fired power plant in a specific time period in one hour, so it can be checked in advance whether the ammonia water in the ammonia water tank is sufficient, or whether the working system is intact, to ensure that the equipment sprays ammonia normally to eliminate pollution.

[0048] Preferably, the resistance value of resistor R is 4.K, and the resistance values ​​of resistor RN of the first, second, third, fourth and fifth control circuits are 1.1K, 1.4M, 1.7M, 2M and 2.3M respectively; the model of NPN transistor QN is 9013; and the model of relay KN is DC12V.

[0049] Example 2

[0050] The method for predicting nitrogen oxide emissions and spraying ammonia based on fusion artificial intelligence of embodiment 2 of the present invention adopts the described control system for predicting nitrogen oxide emissions and spraying ammonia based on fusion artificial intelligence, and includes the following steps: after the AC power supply enters the power input end of the voltage-stabilized power supply A1, the voltage-stabilized power supply A1 will output stable DC power from pins 3 and 4 under the action of the internal circuit to enter the multi-channel control circuit, the data sending circuit and the power input end of the nitrogen oxide detector A2, and each circuit and the detector will be powered on and work; after the nitrogen oxide detector A2 is powered on and works, pin 3 of the signal output end will output analog voltage signals of different sizes according to the different nitrogen oxide concentrations on site and enter the other end of the resistor R1 and the other end of the resistor RN. After the single-chip microcomputer module A3 inputs the dynamically changing analog voltage signal, it will convert the analog voltage signal into a digital signal, and then send it out through the GPRS module A4 via the wireless mobile network. After the mobile phone or PC of the relevant personnel who has established a connection with the GPRS module A4 at the remote end receives the data, the display interface can intuitively display the nitrogen oxide concentration data on site.

[0051] Furthermore, five solenoid valves are coordinated and controlled by five control circuits. The nitrogen oxide detector A2 outputs a dynamically changing analog voltage signal according to the nitrogen oxide concentration at site. When the nitrogen oxide concentration data at site is very small, the voltage signal enters the NPN transistor QN base voltage greater than 0.7V after being reduced and limited by the resistor RN of the first control circuit, and the voltage entering the resistor RN of the second to fifth control circuits is lower than 0.7V. At this time, the NPN transistor QN is turned on and the collector outputs a low level which enters the negative power input terminal of the relay KN. The relay KN is energized to close its control power input terminal and the normally open contact terminal, and then the first solenoid valve is energized to open. Due to the small water flow, the amount of mist ammonia water sprayed from multiple nozzles is relatively small.

[0052] When the on-site nitrogen oxide concentration data is small, the voltage signal enters the NPN transistor QN base voltage greater than 0.7V after being reduced and limited by the resistor RN of the first control circuit and the second control circuit, and the voltage entering the resistor RN of the third to fifth control circuits is lower than 0.7V. At this time, the collector of the NPN transistor QN is turned on and the output of the low level enters the negative power input terminal of the relay KN. The relay KN is energized to close its control power input terminal and the normally open contact terminal, and then the first solenoid valve and the second solenoid valve are energized to open. Due to the slightly larger water flow, the amount of mist ammonia water sprayed from multiple nozzles becomes relatively larger.

[0053] When the on-site nitrogen oxide concentration data is large, the voltage signal enters the NPN transistor QN base voltage greater than 0.7V after being reduced and limited by the resistor RN of the first, second and third control circuits, and the voltage entering the resistor RN of the fourth and fifth control circuits is lower than 0.7V. At this time, the collector of the NPN transistor QN is turned on and the output of the low level enters the negative power input terminal of the relay KN. The relay KN is energized to close its control power input terminal and the normally open contact terminal, and then the first, second and third solenoid valves are energized to open. Due to the large water flow, the amount of mist ammonia water sprayed from multiple nozzles becomes relatively large.

[0054] When the on-site nitrogen oxide concentration data is very large, the voltage signal enters the NPN transistor QN base voltage greater than 0.7V after being reduced and limited by the resistor RN of the first, second, third and fourth control circuits, and the voltage entering the resistor RN of the fifth control circuit is lower than 0.7V. At this time, the NPN transistor QN is turned on and the collector outputs a low level and enters the negative power input terminal of the relay KN. The relay KN is energized and its control power input terminal and the normally open contact terminal are closed, and then the first, second, third and fourth solenoid valves are energized to open. Due to the large water flow, the amount of mist ammonia water sprayed from multiple nozzles becomes relatively more.

[0055] When the on-site nitrogen oxide concentration data is the largest, the voltage signal enters the NPN transistor QN base voltage greater than 0.7V after being reduced and limited by the resistor RN of the first, second, third, fourth and fifth control circuits. At this time, the NPN transistor QN is turned on and the collector outputs a low level and enters the negative power input terminal of the relay KN. The relay KN is energized and its control power input terminal and the normally open contact terminal are closed, and then the first, second, third, fourth and fifth solenoid valves are all energized and opened. Since the water flow is the largest, the amount of mist ammonia water sprayed from multiple nozzles is also at the maximum.

[0056] The embodiments described above are only descriptions of the preferred modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. Based on the fusion artificial intelligence nitrogen oxide emission prediction and ammonia injection control system, it is characterized by: It includes a solenoid valve, an ammonia water tank, a spray tube, a voltage-stabilized power supply, a nitrogen oxide detector, a data transmission circuit, a control circuit, a data acquisition unit, a data analysis unit, a database unit, a data display unit, a data alarm unit and a prediction unit; a plurality of the solenoid valves are installed at the lower end of the ammonia water tank, the other ends of the plurality of the solenoid valves are respectively connected to one end of the spray tube, a plurality of nozzles are installed at the other end of the spray tube, the voltage-stabilized power supply, the data transmission circuit and the control circuit are installed in an electric control box, the nitrogen oxide detector is installed in a monitoring area, the nitrogen oxide detector detects the nitrogen oxide data on site in real time and outputs different voltage signals in real time according to the concentration of nitrogen oxides into the signal input ends of the multiple control circuits and the data transmission circuit, the multiple control circuits The circuits are electrically connected to the power input ends of the plurality of solenoid valves respectively, so as to respectively connect different numbers of the solenoid valves to obtain electricity, thereby adjusting the flow rate of ammonia water sprayed from the nozzle; the data acquisition unit receives and stores the data sent by the data sending circuit, the data analysis unit analyzes the data based on a fusion artificial intelligence algorithm, the database unit stores historical data on the emission time of nitrogen oxides at the production site and the emission amount at the corresponding time, the data display unit displays the specific nitrogen oxide emission data at the site on a display screen in real time, the data alarm unit can generate an alarm signal to prompt on-site personnel when the nitrogen oxides at the site exceed the standard, and the prediction unit can use a pre-trained AI model based on the historical data stored in the database unit to issue a warning that the nitrogen oxides may exceed the standard in advance in the corresponding time period.

2. The fusion-based artificial intelligence nitrogen oxide emission prediction and ammonia injection control system according to claim 1 is characterized in that: Each control circuit includes an electrically connected resistor RN, an NPN transistor and a relay, wherein the positive power input terminal of the relay is connected to the control power input terminal, the collector of the NPN transistor is connected to the negative power input terminal of the relay, and one end of the resistor RN is connected to the base of the NPN transistor.

3. The fusion-based artificial intelligence nitrogen oxide emission prediction and ammonia injection control system according to claim 2 is characterized in that: The data sending circuit includes an electrically connected single-chip microcomputer module, a GPRS module and a resistor R1, the power input end of the single-chip microcomputer module is connected to the power input end of the GPRS module, the signal output end of the single-chip microcomputer module is connected to the signal input end of the GPRS module, and one end of the resistor R1 is electrically connected to the signal input end of the single-chip microcomputer module.

4. The fusion-based artificial intelligence nitrogen oxide emission prediction and ammonia injection control system according to claim 3 is characterized in that: The power input terminals 1 and 2 of the voltage-stabilized power supply are electrically connected to the two poles of the AC 220V power supply, respectively; the power output terminals 3 and 4 of the voltage-stabilized power supply are electrically connected to the positive power input terminal of the relay, the emitter of the NPN transistor, the 1 and 2 pins of the single-chip microcomputer module, and the power input terminals 1 and 2 of the nitrogen oxide detector, respectively; the signal output terminal 3 of the nitrogen oxide detector is electrically connected to the other end of the resistor R1 and the other end of the resistor RN, respectively; the normally open contact terminal of the relay and the emitter of the NPN transistor are electrically connected to the power input terminals of the multiple solenoid valves, respectively.

5. The fusion-based artificial intelligence nitrogen oxide emission prediction and ammonia injection control system according to claim 1 is characterized in that: There are five solenoid valves, and the control circuit is provided with five circuits. The five control circuits coordinately control the five solenoid valves to work and have the same working principle.

6. The fusion-based artificial intelligence nitrogen oxide emission prediction and ammonia injection control system according to claim 5 is characterized in that: The resistance value of the resistor R is 4K, and the resistance values ​​of the resistors RN of the first, second, third, fourth and fifth control circuits are 1.1K, 1.4M, 1.7M, 2M and 2.3M respectively.

7. The fusion-based artificial intelligence nitrogen oxide emission prediction and ammonia injection control system according to claim 1 is characterized in that: The data alarm unit can push text messages to the mobile phones of remote personnel when the nitrogen oxides on site exceed the standard.

8. The fusion-based artificial intelligence nitrogen oxide emission prediction and ammonia injection control system according to any one of claims 1 to 7, characterized in that: The prediction unit is based on the multi-dimensional historical time series collected by the database unit, and after dimensionality reduction processing, selects 127 valid data as input to build a multi-data input fusion short-term prediction model that combines traditional machine learning KNN algorithm and deep learning neural network. The model can accurately predict the concentration of nitrogen oxides in different prediction time periods and realize quantitative control of the ammonia injection flow rate based on this data.

9. A method for predicting nitrogen oxide emissions and injecting ammonia based on fusion artificial intelligence, characterized in that: The integrated artificial intelligence nitrogen oxide emission prediction and ammonia injection control system according to any one of claims 1 to 8 includes the following steps: after the AC power supply enters the power input terminal of the voltage-stabilized power supply, the voltage-stabilized power supply will output stable DC power from pins 3 and 4 under the action of the internal circuit to enter the multi-channel control circuit, the data transmission circuit and the power input terminal of the nitrogen oxide detector, and each circuit and the detector will be powered on; after the nitrogen oxide detector is powered on, pin 3 of the signal output terminal will output analog voltage signals of different sizes according to the different nitrogen oxide concentrations on site, and enter the other end of the resistor R1 and the other end of the resistor RN. After the single-chip microcomputer module inputs the dynamically changing analog voltage signal, it will convert the analog voltage signal into a digital signal, and then send it out through the GPRS module via the wireless mobile network. After the mobile phone or PC of the relevant personnel who has established a remote connection with the GPRS module receives the data, the display interface can intuitively display the nitrogen oxide concentration data on site.

10. The method for predicting nitrogen oxide emissions and injecting ammonia based on fusion artificial intelligence according to claim 9 is characterized in that: The five solenoid valves are controlled by five control circuits. The nitrogen oxide detector outputs a dynamically changing analog voltage signal according to the nitrogen oxide concentration on site. When the nitrogen oxide concentration data on site is very small, the voltage signal enters the NPN transistor base voltage greater than 0.7V after the voltage reduction and current limiting of the resistor RN of the first control circuit, and the voltage of the resistor RN entering the second to fifth control circuits is lower than 0.7V. At this time, the NPN transistor conducts and the collector outputs a low level to the negative power input terminal of the relay. The relay is energized to close its control power input terminal and the normally open contact terminal, and then the first solenoid valve is energized to open. Due to the small water flow, the amount of mist ammonia water sprayed from multiple nozzles is relatively small. When the nitrogen oxide concentration data on site is small, the voltage signal enters the NPN transistor base voltage greater than 0.7V after the voltage reduction and current limiting of the resistor RN of the first control circuit and the second control circuit, and the voltage of the resistor RN entering the third to fifth control circuits is lower than 0.7V. At this time, the collector of the NPN transistor is turned on and the output of the low level enters the negative power input terminal of the relay. The relay is energized to close its control power input terminal and the normally open contact terminal, and then the first solenoid valve and the second solenoid valve are energized to open. Due to the slightly larger water flow, the amount of mist ammonia water sprayed by multiple nozzles becomes relatively larger; When the nitrogen oxide concentration data on site is relatively large, the voltage signal enters the base voltage of the NPN transistor after the voltage reduction and current limiting of the resistor RN of the first, second and third control circuits, and the voltage of the resistor RN entering the fourth and fifth control circuits is greater than 0.7V, and the voltage of the resistor RN entering the fourth and fifth control circuits is lower than 0.7V. At this time, the collector of the NPN transistor is turned on and the output of the low level enters the negative power input terminal of the relay. The relay is energized to close its control power input terminal and the normally open contact terminal, and then the first, second and third solenoid valves are energized to open. Due to the large water flow, the amount of mist ammonia water sprayed by multiple nozzles becomes relatively large; When the nitrogen oxide concentration data on site is very large, the voltage signal enters the NPN transistor base voltage greater than 0.7V after the voltage reduction and current limiting of the resistor RN of the first, second, third and fourth control circuits, and the voltage of the resistor RN entering the fifth control circuit is lower than 0.7V. At this time, the collector of the NPN transistor is turned on and the output of the low level enters the negative power input terminal of the relay. The relay is energized to close its control power input terminal and the normally open contact terminal, and then the first, second, third and fourth solenoid valves are energized to open. Due to the large water flow, the amount of mist ammonia water sprayed by multiple nozzles becomes relatively more. When the on-site nitrogen oxide concentration data is the largest, the voltage signal enters the NPN transistor base voltage greater than 0.7V after being reduced and limited by the resistor RN of the first, second, third, fourth and fifth control circuits. At this time, the NPN transistor is turned on and the collector outputs a low level and enters the negative power input terminal of the relay. The relay is energized and its control power input terminal and normally open contact terminal are closed, and then the first, second, third, fourth and fifth solenoid valves are all energized and opened. Since the water flow is the largest, the amount of mist ammonia water sprayed from multiple nozzles is also at the maximum.

Images