Intelligent control method and system for multi-tower desulfurization spray device

Abstract

The present application provides an intelligent control method and system for a multi-tower desulfurization spray device, which relates to the field of spray control technology, including: the multi-tower desulfurization spray device includes multiple spray pumps corresponding to multiple biological desulfurization towers, and monitors multiple biological desulfurization towers; makes control decisions on multiple spray pumps respectively; sets multivariate evaluation factors for desulfurization spray; builds a multi-layer optimization channel for desulfurization spray; conducts multi-level joint optimization on multiple desulfurization spray decision spaces respectively; and sprays multiple biological desulfurization towers according to the multi-tower desulfurization spray device based on multiple desulfurization spray optimization strategies. The present application can solve the technical problem in the prior art that the spray control parameters of the desulfurization pump fail to fully adapt to the needs of specific working conditions, resulting in unbalanced desulfurization spraying and further affecting the desulfurization efficiency. By constructing a multi-layer optimization channel, multi-level joint optimization of spray parameters is achieved, the spray strategy of each tower is optimized, and the desulfurization efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of spray control technology, and in particular to an intelligent control method and system for a multi-tower desulfurization spray device. Background Art

[0002] Multi-tower desulfurization spray devices are widely used in industrial waste gas treatment, mainly removing sulfur dioxide from waste gas through the biodegradation process. In the biological desulfurization tower, specially cultivated bacterial strains use sulfur dioxide in the waste gas as an energy source to carry out biological conversion reactions, thereby achieving desulfurization. In addition to the control of waste gas flow and composition, it is also necessary to regularly and quantitatively supplement the bacteria in the desulfurization tower with nutrient solution to maintain the activity of the bacteria and the desulfurization efficiency. The number of biological desulfurization towers configured for each replenishment of nutrient solution varies depending on the front-end biogas production. In existing desulfurization spray devices, the spray pump is often adjusted based on some fixed parameters (such as flow, pressure, etc.), but the dynamic changes of waste gas, such as fluctuations in waste gas flow, temperature, composition, and even the gas-liquid contact state in the tower, are not fully considered, resulting in uneven spraying, too much absorption liquid in some towers, resulting in waste; while other towers have too little spray liquid, which cannot fully react, reducing the desulfurization efficiency and even leading to long-term equipment loss.

[0003] In summary, the prior art has a technical problem in that the spray control parameters of the desulfurization pump fail to fully adapt to the requirements of specific working conditions, resulting in unbalanced desulfurization spraying, which further affects the desulfurization efficiency. Summary of the invention

[0004] The purpose of this application is to provide an intelligent control method and system for a multi-tower desulfurization spray device, so as to solve the technical problem in the prior art that the spray control parameters of the desulfurization pump fail to fully adapt to the requirements of specific working conditions, resulting in unbalanced desulfurization spraying and further affecting the desulfurization efficiency.

[0005] In view of the above problems, the present application provides an intelligent control method and system for a multi-tower desulfurization spray device.

[0006] In the first aspect, the present application provides an intelligent control method for a multi-tower desulfurization spray device, which is implemented by an intelligent control system of the multi-tower desulfurization spray device, wherein the intelligent control method for the multi-tower desulfurization spray device includes: the multi-tower desulfurization spray device includes a plurality of spray pumps corresponding to a plurality of biological desulfurization towers; the plurality of biological desulfurization towers are monitored to obtain a plurality of desulfurization tower monitoring data; according to the plurality of desulfurization tower monitoring data, control decisions are made on the plurality of spray pumps respectively to establish a plurality of desulfurization spray decision spaces ; Set up a desulfurization spray multivariate evaluation factor, wherein the desulfurization spray multivariate evaluation factor includes spray balance, spray sufficiency and spray energy efficiency; build a desulfurization spray multi-layer optimization channel according to the desulfurization spray multi-layer evaluation factor; based on the monitoring data of the multiple desulfurization towers, according to the desulfurization spray multi-layer optimization channel, perform multi-level joint optimization on the multiple desulfurization spray decision spaces respectively, and obtain multiple desulfurization spray optimization strategies; based on the multiple desulfurization spray optimization strategies, spray the multiple biological desulfurization towers according to the multi-tower desulfurization spray device.

[0007] In the second aspect, the present application also provides an intelligent control system for a multi-tower desulfurization spray device, which is used to execute the intelligent control method for a multi-tower desulfurization spray device as described in the first aspect, wherein the intelligent control system for the multi-tower desulfurization spray device includes: a monitoring data acquisition module, which is used to monitor the multiple biological desulfurization towers and obtain multiple desulfurization tower monitoring data; a decision space establishment module, which is used to make control decisions on the multiple spray pumps respectively according to the multiple desulfurization tower monitoring data, and establish multiple desulfurization spray decision spaces; an evaluation factor setting module, which is used to set the desulfurization spray multivariate evaluation factor, Among them, the desulfurization spray multivariate evaluation factors include spray balance, spray sufficiency and spray energy efficiency; an optimization channel building module, the optimization channel building module is used to build a desulfurization spray multi-layer optimization channel according to the desulfurization spray multivariate evaluation factors; a joint optimization module, the joint optimization module is used to perform multi-level joint optimization on the multiple desulfurization spray decision spaces based on the multiple desulfurization tower monitoring data and the desulfurization spray multi-layer optimization channel to obtain multiple desulfurization spray optimization strategies; a strategy execution module, the strategy execution module is used to spray the multiple biological desulfurization towers according to the multi-tower desulfurization spray device based on the multiple desulfurization spray optimization strategies.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] The multi-tower desulfurization spray device includes multiple spray pumps corresponding to multiple biological desulfurization towers; the multiple biological desulfurization towers are monitored to obtain multiple desulfurization tower monitoring data; according to the multiple desulfurization tower monitoring data, control decisions are made for the multiple spray pumps respectively, and multiple desulfurization spray decision spaces are established; a desulfurization spray multivariate evaluation factor is set, wherein the desulfurization spray multivariate evaluation factor includes spray balance, spray margin and spray energy efficiency; according to the desulfurization spray multivariate evaluation factor, a desulfurization spray multi-layer optimization channel is built; based on the multiple desulfurization tower monitoring data and according to the desulfurization spray multi-layer optimization channel, the multiple desulfurization spray decision spaces are respectively optimized at multiple levels to obtain multiple desulfurization spray optimization strategies; based on the multiple desulfurization spray optimization strategies, the multiple biological desulfurization towers are sprayed according to the multi-tower desulfurization spray device. In other words, a decision space is established for each desulfurization tower through monitoring data, multi-source evaluation factors are set, a multi-layer optimization channel is constructed, multi-level joint optimization is performed on multiple decision spaces, the operation of each spray pump is adjusted, and the optimal spraying of multiple biological desulfurization towers is achieved, thereby improving the desulfurization efficiency.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0012] Figure 1 This is a flow chart of the intelligent control method of the multi-tower desulfurization spray device of this application;

[0013] Figure 2 This is a schematic diagram of the structure of the intelligent control system of the multi-tower desulfurization spray device of this application.

[0014] Explanation of the accompanying drawings: monitoring data collection module 11, decision space establishment module 12, evaluation factor setting module 13, optimization channel construction module 14, joint optimization module 15, strategy execution module 16. DETAILED DESCRIPTION

[0015] This application solves the technical problem in the prior art that the spray control parameters of the desulfurization pump fail to fully adapt to the needs of specific working conditions, resulting in unbalanced desulfurization spraying and further affecting the desulfurization efficiency by providing an intelligent control method and system for a multi-tower desulfurization spraying device. A decision space is established for each desulfurization tower through monitoring data, multi-source evaluation factors are set, a multi-layer optimization channel is constructed, and multi-level joint optimization is performed on multiple decision spaces to adjust the operation of each spray pump, so as to achieve optimal spraying of multiple biological desulfurization towers and improve the desulfurization efficiency.

[0016] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0017] For example, please refer to the attached Figure 1 The present application provides an intelligent control method for a multi-tower desulfurization spray device, wherein the intelligent control method for a multi-tower desulfurization spray device is applied to an intelligent control system of the multi-tower desulfurization spray device, and the intelligent control method for the multi-tower desulfurization spray device specifically includes the following steps:

[0018] Step 1: The multi-tower desulfurization spray device includes a plurality of spray pumps corresponding to a plurality of biological desulfurization towers, and the plurality of biological desulfurization towers are monitored to obtain monitoring data of a plurality of desulfurization towers.

[0019] Specifically, a multi-tower desulfurization spray device refers to a desulfurization device consisting of multiple biological desulfurization towers and corresponding multiple spray pumps. Each biological desulfurization tower is responsible for treating a certain amount of waste gas, and the spray pump is responsible for spraying liquid, usually nutrient solution, into the desulfurization tower to support the growth of bacteria in the tower and improve the desulfurization efficiency. The multi-tower system can treat multiple waste gas streams at the same time and is suitable for large-scale desulfurization tasks. The biological desulfurization tower is a device that converts and removes sulfur dioxide in the waste gas through microorganisms (such as specific bacterial strains). The bacteria use sulfur dioxide as energy to complete the biological desulfurization reaction. The spray pump is a liquid delivery device. In the biological desulfurization tower, the spray pump is responsible for spraying the nutrient solution into the tower to ensure that the bacteria can obtain the required nutrients. The spray pump accurately controls the amount of nutrient solution added by controlling parameters such as liquid flow and pressure to help the bacteria remain active.

[0020] The multi-tower desulfurization spray device includes multiple biological desulfurization towers, each of which is equipped with a spray pump. The function of the spray pump is to ensure that sufficient nutrient solution is sprayed into the desulfurization tower to support the growth of bacterial strains. In order to ensure the efficient operation of the biological desulfurization system, it is necessary to regularly monitor the working status of multiple biological desulfurization towers and the health status of the bacterial strains in the tower to ensure that the spraying amount of nutrient solution can be adjusted according to actual needs.

[0021] Each biological desulfurization tower will be equipped with sensors to monitor environmental data such as exhaust gas flow, temperature, humidity, etc., and will also monitor the activity of bacteria in the tower. For example, optical sensors are used to monitor the reproduction status of bacteria in real time, or pH sensors are used to monitor the pH of the desulfurization liquid, so as to determine whether the growth environment of bacteria is suitable. The monitoring data, that is, the monitoring data of multiple desulfurization towers, are obtained in real time through sensors, including various parameters of the desulfurization tower operation status, such as pH value, temperature, humidity, SO2 concentration, etc., as well as the status of bacterial strains, such as the number of bacteria, activity or other biological indicators, to ensure that the bacteria can work in the best state. Through the monitoring data, the amount of spray liquid is dynamically adjusted according to the status of the bacteria in the tower and the exhaust gas flow, to ensure that the bacteria in each tower obtains enough nutrient solution to maintain its activity and desulfurization efficiency.

[0022] Step 2: Based on the monitoring data of the multiple desulfurization towers, control decisions are made on the multiple spray pumps respectively to establish multiple desulfurization spray decision spaces.

[0023] Specifically, according to the monitoring data of multiple desulfurization towers, according to the monitoring data of each desulfurization tower, the spray control retrieval target and spray control retrieval constraint are set for each tower. According to the monitoring data of each desulfurization tower and the set retrieval target and constraint, the corresponding control scheme is retrieved for each spray pump, involving the adjustment of parameters such as the flow rate, pressure, and working time of the spray pump. According to the control scheme of each desulfurization tower, the spray control interval is identified, that is, the control interval of each parameter, which jointly affects the control of the spray pump, and helps to find the appropriate operating range for each desulfurization tower under different conditions. And based on this, the spray control trigger domain is determined, that is, the timing or condition range for starting or adjusting each spray pump. According to the control decision of each desulfurization tower, a desulfurization spray decision space is generated for each tower, which contains all possible spray control decisions under different working conditions. Each decision space will contain appropriate spray operation plans based on the specific data and needs of the tower.

[0024] The spray control decision spaces of all desulfurization towers will be integrated into an overall spray decision system, which can select the most appropriate spray control strategy from its corresponding decision space according to the real-time monitoring data of each desulfurization tower and automatically execute it. By providing tailor-made spray control decisions for each desulfurization tower, it is ensured that each tower can get the most suitable spray operation under different working conditions, and the working state of the spray pump is dynamically adjusted according to the actual situation of different desulfurization towers, so as to achieve fine control of multiple desulfurization towers and spray pumps and improve desulfurization efficiency.

[0025] Step three: setting the desulfurization spray multivariate evaluation factor, wherein the desulfurization spray multivariate evaluation factor includes spray balance, spray margin and spray energy efficiency.

[0026] Specifically, the real-time monitoring system is used to collect the operating data of each desulfurization tower, including spray volume, bacterial activity index, exhaust gas desulfurization rate, spray liquid flow rate, etc. At the same time, the energy consumption data of the spray pump, such as power consumption, is recorded. The multivariate evaluation factors of desulfurization spray are set to analyze the effect of spray operation from multiple angles and comprehensively evaluate the performance of the desulfurization spray device. The multivariate evaluation factors of desulfurization spray include spray balance, spray adequacy and spray energy efficiency. Among them, the spray balance reflects the balance of spray distribution between each biological desulfurization tower, which can be measured by calculating the standard deviation or coefficient of variation of the spray volume of each desulfurization tower. If some towers are not sprayed enough and other towers are sprayed too much, the balance is low; the spray adequacy indicates whether the supply of spray liquid meets the needs of bacterial strains in the biological desulfurization tower. It is judged by comparing the actual spray volume with the theoretical spray volume. It is usually a value between 0 and 1. The closer the adequacy is to 1, the more the supply of spray liquid meets the demand, which helps to maintain an efficient desulfurization reaction. The spraying energy efficiency measures the energy utilization efficiency of the spraying operation. It is calculated by calculating the efficiency per unit spraying volume. It refers to whether the energy consumption of the spraying operation is in the optimal state while achieving the desulfurization target. High energy efficiency means that a higher desulfurization efficiency is achieved under unit energy consumption. Combining the results of the above three factors, the desulfurization spraying operation is evaluated in multiple dimensions. By monitoring and adjusting the spraying volume, the rationality of the distribution of spraying resources among the desulfurization towers is improved, excessive or insufficient spraying is avoided, and the efficiency and energy efficiency of the spraying system are improved.

[0027] Step 4: According to the desulfurization spray multi-evaluation factors, a desulfurization spray multi-layer optimization channel is established.

[0028] Specifically, the goal of desulfurization spraying is to comprehensively optimize these multiple evaluation factors so that efficient, energy-saving and balanced spraying effects can be achieved under various working conditions. A multi-layer optimization channel is established to optimize the spraying control strategy, and target optimization is performed from multiple dimensions and levels. Preliminary optimization is performed for the goals of each biological desulfurization tower to ensure that various control goals can be met in a local range, including spray balance, sufficiency, and energy efficiency. By adjusting the flow rate, spray angle and other parameters of the spray pump, the spray liquid in each desulfurization tower is evenly distributed; according to the actual needs of the desulfurization tower and the requirements of the bacterial strain, the running time and liquid flow rate of the spray pump are adjusted to ensure that the bacteria in each desulfurization tower get enough nutrient solution; by finely controlling the start and shutdown time of the spray pump, unnecessary energy consumption is reduced. Taking into account multiple factors such as spray balance, sufficiency and energy efficiency, optimization is performed globally through an integrated optimization algorithm to achieve the optimal balance of the spraying effect and meet the specific needs of each desulfurization tower.

[0029] According to the multivariate evaluation factors of desulfurization spraying, the spray balance prediction branch, spray sufficiency prediction branch, and spray energy efficiency prediction branch are constructed according to the machine learning model, and the three prediction branches are connected as parallel nodes to obtain the desulfurization spray prediction channel. By setting control constraints (such as the expected intervals of balance, sufficiency, and energy efficiency), a desulfurization spray constraint optimization channel is constructed to ensure that the spray control is within a reasonable range, avoid excessive or insufficient spray liquid supply, and ensure desulfurization efficiency. According to the importance evaluation of each evaluation factor in the multivariate evaluation factors of desulfurization spraying, the optimization decision is made, the weight of each evaluation factor is determined, and the operation of the spray tower is optimized through these weights to construct the desulfurization spray fitness optimization channel. The desulfurization spray prediction channel, the desulfurization spray constraint optimization channel, and the desulfurization spray fitness optimization channel are connected to obtain the desulfurization spray multi-layer optimization channel, which comprehensively considers prediction, constraints, and fitness to provide a more comprehensive and accurate desulfurization spray control strategy. By building a multi-layer optimization channel for desulfurization spraying, the spraying balance, abundance and energy efficiency are balanced, the overall performance is optimized and the desulfurization effect is improved.

[0030] Step 5: Based on the monitoring data of the multiple desulfurization towers and according to the desulfurization spray multi-layer optimization channel, the multiple desulfurization spray decision spaces are respectively optimized at multiple levels to obtain multiple desulfurization spray optimization strategies.

[0031] Specifically, the desulfurization spray multi-layer optimization channel, for the spray decision space of each desulfurization tower, combines the monitoring data of multiple desulfurization towers to perform multi-level joint optimization to obtain the best desulfurization spray optimization strategy. Multi-level joint optimization refers to the use of different evaluation factors (such as spray balance, sufficiency, energy efficiency, etc.) in the multi-level decision space to screen and optimize layer by layer to ensure that each spray decision not only satisfies the constraints locally, but also achieves the best optimization effect overall. The following steps are performed for each desulfurization spray decision space in the multiple desulfurization spray decision spaces. All decision plans for each desulfurization spray decision space and the corresponding desulfurization tower monitoring data are input into the desulfurization spray prediction channel in the desulfurization spray multi-layer optimization channel to obtain prediction results, including predicted spray balance, predicted spray sufficiency and predicted spray energy efficiency.

[0032] The prediction results are input into the desulfurization spray constraint optimization channel in the desulfurization spray multi-layer optimization channel, and the predicted spray balance, predicted spray sufficiency and predicted spray energy efficiency all meet the constraints, that is, the decision schemes that meet the expected interval are selected to form multiple desulfurization spray optimization spaces corresponding to multiple desulfurization spray decision spaces. Multiple desulfurization spray decision spaces correspond to multiple desulfurization spray optimization spaces one by one. Multiple desulfurization spray optimization spaces are input into the desulfurization spray fitness optimization channel in the desulfurization spray multi-layer optimization channel, and the spray decision in each desulfurization spray optimization space is maximized and optimized. According to the spray balance weight, spray sufficiency weight and spray energy efficiency weight, multiple desulfurization spray optimization spaces are optimized to find the best desulfurization spray control strategy, that is, multiple desulfurization spray optimization strategies. Multiple desulfurization spray optimization strategies are added to multiple desulfurization spray optimization strategies to obtain the spray decision that best suits each desulfurization tower. Through multi-level joint optimization based on multiple desulfurization tower monitoring data and desulfurization spray multi-layer optimization channels, the spray decision of each desulfurization tower is optimized to ensure that the spray decision of each desulfurization tower achieves the best balance under different indicators (balance, redundancy, energy efficiency), thereby improving the overall desulfurization efficiency.

[0033] Step six: Based on the multiple desulfurization spray optimization strategies, the multiple biological desulfurization towers are sprayed according to the multi-tower desulfurization spray device.

[0034] Specifically, multiple desulfurization spray optimization strategies are used to control the actual spraying of multiple biological desulfurization towers. Each desulfurization spray optimization strategy is based on different spray decision spaces and obtains the optimal result through multi-level optimization steps (such as balance, sufficiency and energy efficiency). The final spray control scheme will comprehensively consider the optimization results of various evaluation factors, thereby effectively improving the operating efficiency of the entire desulfurization system. According to multiple desulfurization spray optimization strategies, the working parameters of multiple spray pumps are adjusted, and the spray pumps are started or adjusted to accurately spray the nutrient solution on each biological desulfurization tower. The spraying operation of each desulfurization tower will operate within the set spray balance, sufficiency and energy efficiency range to ensure the stability of the spraying effect and maximize the desulfurization efficiency. In the actual operation process, the multi-tower desulfurization spray device will dynamically adjust the spraying amount and spraying mode according to real-time monitoring data. For example, when the gas flow rate of the desulfurization tower increases, the spraying amount is increased according to the spray decision strategy to maintain the spray balance. At the same time, if the monitoring spray effect deviates from the expectation (such as the balance, sufficiency or energy efficiency decreases), the spray decision will be re-evaluated through the feedback mechanism and necessary adjustments will be made. By optimizing the spray decision, it is ensured that the spray effect of each desulfurization tower meets the optimal standard, and resources are allocated according to the actual situation of each desulfurization tower, so that the resource use of each tower is more efficient, and the precise control of the spray liquid and the optimal utilization of resources are achieved, thereby improving the desulfurization efficiency and reducing resource waste.

[0035] Further, step 2 of this application includes:

[0036] Traverse the multiple biological desulfurization towers and extract the first biological desulfurization tower; extract the first desulfurization tower monitoring data corresponding to the first biological desulfurization tower based on the multiple desulfurization tower monitoring data; use the first desulfurization tower monitoring data as the spray control retrieval constraint; use the control parameters of the spray pump corresponding to the first desulfurization tower monitoring data as the spray control retrieval target; perform spray control record retrieval according to the spray control retrieval constraint and the spray control retrieval target to obtain a spray control scheme record set; perform multi-dimensional spray control interval identification according to the spray control scheme record set to generate a spray control trigger domain; use the first desulfurization tower monitoring data as the spray control target, make a control decision on the spray control trigger domain, generate a first desulfurization spray decision space, and add the first desulfurization spray decision space to the multiple desulfurization spray decision spaces.

[0037] Specifically, all biological desulfurization towers are traversed, and the monitoring data of each desulfurization tower is extracted one by one. Taking the first biological desulfurization tower as an example, a biological desulfurization tower is arbitrarily selected from multiple biological desulfurization towers as the first biological desulfurization tower, and the first desulfurization tower monitoring data corresponding to the first biological desulfurization tower is extracted, including the exhaust gas flow rate, temperature and humidity, pH value and other key parameters of the tower, as well as the bacterial species inside the first biological desulfurization tower. The monitoring data of the first desulfurization tower is used as the spray control retrieval constraint, that is, the set constraint condition, which is used to limit the working parameter range of the spray pump. The control parameters of the spray pump corresponding to the monitoring data of the first desulfurization tower are used as the spray control retrieval target, including the flow rate, pressure, frequency and other control parameters of the spray pump, to ensure that the spray pump can provide the appropriate amount of nutrient solution as needed.

[0038] Spray control record retrieval is performed according to the spray control retrieval constraints and spray control retrieval targets. Past spray control records can be retrieved using a database or record storage system to obtain a spray control scheme record set, including multiple control schemes that match the current constraints and targets, corresponding to different operating conditions and control strategies, including the control parameters (such as flow rate, spray volume, etc.) of the spray pump under different operating conditions (such as different temperatures, humidity, exhaust gas flow, etc.).

[0039] According to the data in the spray control scheme record set, determine the control dimensions that need to be considered, including temperature, humidity, exhaust gas flow, pH value, etc. These factors will affect the control parameters such as flow and pressure of the spray pump. Analyze the spray control scheme record set to identify the spray pump control scheme under different working conditions (such as different temperature and humidity conditions). According to different working conditions (such as temperature, humidity, etc.), identify multiple control intervals from the spray control scheme. Multidimensional spray control interval refers to the spray control range or interval based on multiple variables (such as temperature, humidity, exhaust gas flow, etc.), including control intervals of different parameters. Further determine the triggering conditions for the start of the spray pump and generate the spray control trigger domain, that is, determine the timing or condition range for the start or adjustment of the spray pump. When the real-time monitoring data of the biological desulfurization tower meets the triggering conditions, the spray pump will be triggered. The triggering domain includes not only the start-up conditions of the spray pump, but also the working status of the spray pump (such as flow, pressure, etc.) and time period.

[0040] According to the monitoring data of the first desulfurization tower, the spray control target is set. According to the set control target and the spray control trigger domain, the spray control decision is generated. The control decision will be dynamically adjusted according to the monitoring data of the first desulfurization tower. The spray control decision refers to the specific decision to determine the operation of the spray pump based on the monitoring data and the control target, including the setting of control parameters (such as spray volume, flow rate, duration, etc.) and whether to start the spray pump. Through the analysis of the monitoring data and the formulation of the control decision, the first desulfurization spray decision space is generated, which includes multiple spray decision schemes corresponding to the first biological desulfurization tower, each scheme corresponding to different working conditions. The spray decision space of the first desulfurization tower is added to the multiple desulfurization spray decision spaces. By analogy, the above operations are performed on other biological desulfurization towers in the multiple biological desulfurization towers to obtain the desulfurization spray decision space corresponding to each biological desulfurization tower, and they are added to the multiple desulfurization spray decision spaces together to form a spray decision space set containing all towers.

[0041] Multiple desulfurization spray decision spaces can independently adjust the spray operation of each tower according to the real-time data of each tower. Each desulfurization tower will have a corresponding spray decision space, and different spray control strategies will be formulated based on the operating data of different towers. The operation of the spray pump is adjusted according to the real-time monitoring data of each desulfurization tower to ensure that each desulfurization tower obtains the appropriate amount of nutrient solution according to its specific operating conditions. Through automated control and precise adjustment, the desulfurization effect is improved and energy consumption is reduced.

[0042] Further, step 4 of this application includes:

[0043] Evaluation feature learning is performed according to the desulfurization spray multivariate evaluation factors to establish a desulfurization spray prediction channel; the multiple biological desulfurization towers are expected to be set according to the desulfurization spray multivariate evaluation factors to establish a desulfurization spray constraint optimization channel, wherein the desulfurization spray constraint optimization channel includes multiple desulfurization spray constraints, and each desulfurization spray constraint includes an expected spray balance interval, an expected spray margin interval, and an expected spray energy efficiency interval corresponding to each biological desulfurization tower; importance evaluation is performed according to the desulfurization spray multivariate evaluation factors to obtain the importance evaluation coefficient of each factor; weights are assigned to the desulfurization spray multivariate evaluation factors according to the importance evaluation coefficients of each factor to construct a desulfurization spray fitness optimization channel, wherein the desulfurization spray fitness optimization channel includes a spray balance weight, a spray margin weight, and a spray energy efficiency weight; the desulfurization spray prediction channel, the desulfurization spray constraint optimization channel, and the desulfurization spray fitness optimization channel are connected to generate the desulfurization spray multi-layer optimization channel.

[0044] Specifically, evaluation feature learning is performed based on the desulfurization spray multivariate evaluation factors, and features useful for prediction or classification tasks are automatically learned from the original data to establish a desulfurization spray prediction channel. The desulfurization spray multivariate evaluation factors are combined with machine learning models for training to predict the changing trends of spray balance, sufficiency, and energy efficiency. These evaluation factors are predicted through historical data or real-time data to generate a prediction model.

[0045] According to the multi-evaluation factors of desulfurization spraying, each biodesulfurization tower is expected to be set, including the expected interval of spray balance, spray margin and spray energy efficiency, to ensure that the spraying operation can meet the established performance standards. By setting the expected value range of spray balance, spray margin and spray energy efficiency, the expected value range is defined, which represents the ideal spray control range, ensuring that the spraying system operates within a reasonable performance range.

[0046] For each desulfurization tower, the expected range of spray balance may be different, depending on the scale and gas processing capacity of the tower; spray adequacy may be affected by the current gas flow and biogas production, so the expected range should also be adjusted; the expected range of spray energy efficiency depends on the energy management needs, and the energy efficiency of the system can be controlled by setting the range. By setting the control constraints of the spray system (such as the expected range of balance, adequacy and energy efficiency), a desulfurization spray constraint optimization channel is established.

[0047] The importance of each evaluation factor (spray balance, sufficiency, energy efficiency) is quantified through algorithms (such as information gain, correlation analysis, etc.) to obtain the importance evaluation coefficient of each evaluation factor. According to the importance evaluation coefficient, each evaluation factor is weighted and a desulfurization spray fitness optimization channel is constructed, so that the spray control decision can focus on the factors that have the greatest impact on the desulfurization efficiency. The impact of different factors on the final desulfurization effect is evaluated through historical data or expert evaluation. For example, the relevance and influence of each factor are calculated to evaluate its importance; the impact of each factor is scored based on the experience of experts, and the weight of each factor is obtained by weighting.

[0048] By calculating the correlation between spray balance, adequacy, energy efficiency and desulfurization efficiency, factors with higher correlation will be given higher weights. For example, if spray balance has the greatest impact on desulfurization efficiency, spray balance will be given a higher weight. The importance evaluation coefficient of each factor is assigned to each desulfurization spray multivariate evaluation factor, namely, the spray balance weight, spray adequacy weight and spray energy efficiency weight, to form a complete fitness optimization channel. The goal of the fitness optimization channel is to maximize the desulfurization effect by optimizing the control strategy of spray balance, adequacy and energy efficiency, while avoiding unnecessary waste of resources. According to the weight of each factor, the control objectives of each evaluation factor are weighted to obtain the fitness function.

[0049] The desulfurization spray prediction channel, desulfurization spray constraint optimization channel and desulfurization spray fitness optimization channel are connected to form a complete desulfurization spray multi-layer optimization channel. The desulfurization spray prediction channel is used to provide prediction results of spray balance, sufficiency and energy efficiency; the desulfurization spray constraint optimization channel is used to screen the optimization scheme according to the set expected value and constraint conditions; the desulfurization spray fitness optimization channel is used to make the final selection of the optimization scheme according to the weight of the evaluation factor. Through the constructed multi-layer optimization channel, multiple optimization paths are combined to gradually adjust the control parameters of the spray pump, optimize the spray liquid distribution and energy efficiency of each biological desulfurization tower, adjust the flow of the spray liquid in real time, control the opening and closing of the spray pump, and ensure that each desulfurization tower operates within the optimal range. By optimizing the spray balance, the spray liquid is evenly distributed among different desulfurization towers. By optimizing the spray sufficiency and energy efficiency, energy waste is reduced and the efficiency of the spray tower is improved.

[0050] Furthermore, the present application also includes the following steps:

[0051] According to the desulfurization spray multivariate evaluation factor, the desulfurization spray evaluation record is retrieved to obtain the desulfurization tower monitoring data sample set, the spray decision sample set, the spray balance sample set, the spray margin sample set and the spray energy efficiency sample set; based on K spray evaluation feature learning models, the desulfurization tower monitoring data sample set, the spray decision sample set and the spray balance sample set are integrated and fused to generate a spray balance prediction branch, wherein K is a positive integer greater than 1; based on the ... The tower monitoring data sample set, the spray decision sample set and the spray margin sample set are integrated and fused to generate a spray margin prediction branch; based on the K spray evaluation feature learning models, the desulfurization tower monitoring data sample set, the spray decision sample set and the spray energy efficiency sample set are integrated and fused to generate a spray energy efficiency prediction branch; the spray balance prediction branch, the spray margin prediction branch and the spray energy efficiency prediction branch are connected as parallel nodes to generate the desulfurization spray prediction channel.

[0052] Specifically, according to the multivariate evaluation factors of desulfurization spraying, the desulfurization spraying evaluation records are retrieved from the historical data, and the desulfurization tower monitoring data sample set, spray decision sample set, spray balance sample set, spray adequacy sample set and spray energy efficiency sample set are extracted. The desulfurization tower monitoring data sample set includes various parameters reflecting the operating status of the desulfurization tower, such as pH value, temperature, humidity, SO2 concentration, etc., as well as the status of bacterial strains. The spray decision sample set records the control measures taken during the spraying operation, including control parameters such as spraying volume, spraying frequency, and the opening of the regulating valve. Spray balance measures the balance of the spray volume distribution of each desulfurization tower, spray adequacy measures whether the supply of spray liquid meets the needs of the bacterial strains of the desulfurization tower, and spray energy efficiency measures the desulfurization effect and energy efficiency of the spraying operation under unit spray liquid consumption.

[0053] According to the desulfurization tower monitoring data sample set, spray decision sample set and spray balance sample set, K spray evaluation feature learning models are integrated and fused. Specifically, multiple machine learning models are used for integrated learning, such as decision tree, random forest, support vector machine, neural network, etc., to train multiple spray balance predictors. In order to improve the prediction accuracy, the output results of multiple spray balance predictors are integrated, that is, the output data sets of multiple spray balance predictors are used as input information, and the spray balance sample set is used as output information to obtain the final prediction result, and the spray balance prediction fusion device is trained. Multiple spray balance predictors are merged as parallel independent nodes to generate a spray balance prediction processing layer, and the output layer of the spray balance prediction processing layer is merged with the input layer of the spray balance prediction fusion device to generate a spray balance prediction branch.

[0054] Integrated fusion learning improves the accuracy and robustness of predictions by combining the prediction results of multiple models. The prediction results of K sprinkler evaluation feature learning models are integrated and fused to generate sprinkler balance prediction branches, including voting method, weighted average method, etc. Similarly, the above steps are also performed for the sprinkler adequacy sample set and sprinkler energy efficiency sample set to obtain sprinkler adequacy prediction branches and sprinkler energy efficiency prediction branches. Each branch consists of multiple machine learning models, which are responsible for predicting the state of the sprinkler system from different angles and finally providing a fusion result. The steps are similar to the corresponding sub-rights later, so I will not elaborate on them here.

[0055] The spray balance prediction branch, spray adequacy prediction branch and spray energy efficiency prediction branch are connected in parallel. These three branches will make predictions at the same time and pass the output results to the next level channel to generate a desulfurization spray prediction channel. The spray balance prediction branch, spray adequacy prediction branch and spray energy efficiency prediction branch predict different dimensions of the spray system respectively. Each branch generates the final prediction result through the integrated calculation of multiple machine learning models. The predictions of spray balance, spray adequacy and spray energy efficiency are generated by different model groups respectively, and then the output results of each branch are merged, and the outputs of each prediction branch are combined into a comprehensive evaluation result. Each prediction branch is calculated independently, and then the results of the three branches are merged to finally obtain the predicted value of each spray factor. By processing multiple evaluation factors (spray balance, adequacy, energy efficiency) in parallel, a comprehensive evaluation of the spray system is achieved, and the accuracy of control decisions is improved.

[0056] Furthermore, the present application also includes the following steps:

[0057] According to the desulfurization tower monitoring data sample set, the spray decision sample set and the spray balance sample set, supervised training is performed on the K spray evaluation feature learning models to generate multiple spray balance predictors; the output data sets of the multiple spray balance predictors are used as input information, and the spray balance sample set is used as output information to train a spray balance prediction fusion device; the multiple spray balance predictors are merged as parallel independent nodes to generate a spray balance prediction processing layer; the spray balance prediction processing layer is merged with the input layer of the spray balance prediction fusion device to generate the spray balance prediction branch.

[0058] Specifically, according to the desulfurization tower monitoring data sample set, spray decision sample set and spray balance sample set, K spray evaluation feature learning models are trained through supervised learning to obtain multiple spray balance predictors. Select appropriate machine learning models, such as random forest, support vector machine (SVM), neural network, etc., to train the input data. For each model, optimize its parameters (such as the depth of the decision tree, the number of layers of the neural network, etc.) to ensure that it can accurately predict the spray balance. The spray balance predictor is a prediction tool generated based on machine learning models (such as regression models, support vector machines, etc.) to predict the spray balance. Assume that K different models (K is greater than 1) are selected, such as model 1 is a decision tree, model 2 is a support vector machine, and model 3 is a neural network. First, the data set is preprocessed, including data cleaning and data standardization / normalization processing, to ensure the quality of the data. The data set is divided into a training set and a test set, such as an 80% training set and a 20% validation set.

[0059] Each model is trained using a training set (including a desulfurization tower monitoring data sample set, a spray decision sample set, and a spray balance sample set). Each model learns how to predict spray balance based on input data (desulfurization tower monitoring data and spray decision data). For example, during training, the decision tree automatically selects a feature (such as spray liquid flow) for splitting until the optimal splitting rule is reached. Each split will generate a judgment condition and eventually form a decision tree model; SVM will find the best hyperplane based on different features and target values, and divide the data points into different areas for regression prediction; the neural network is trained through a multi-layer neuron structure, and the weight parameters are optimized using the back propagation algorithm to gradually approach the target value (spray balance).

[0060] Use the test set data to verify the accuracy of the model, calculate its error (such as mean square error MSE, which is the average of the squared difference between the predicted value and the true value), evaluate the generalization ability of the model, and avoid overfitting. According to the evaluation results of the model, adjust the hyperparameters, such as the maximum depth of the decision tree, the kernel function type of SVM, the learning rate of the neural network, etc., to further improve the model performance. Once the training process is completed, K models (such as decision trees, SVMs, neural networks) will generate spray balance predictors respectively. Each model can independently predict the spray balance based on the input data (desulfurization tower monitoring data and spray decision data).

[0061] In order to improve the prediction accuracy, the output results of multiple sprinkler balance predictors are integrated, and the output data sets of multiple sprinkler balance predictors are used as input information, and the sprinkler balance sample set is used as output information to train the sprinkler balance prediction fusion. The results of multiple sprinkler balance predictors are integrated, and the final sprinkler balance prediction value is obtained through a fusion mechanism (such as weighted average, voting mechanism, etc.). The sprinkler balance prediction fusion is an integrated model that combines the outputs of multiple sprinkler balance predictors to produce more accurate prediction results. For example, suppose there are 3 sprinkler balance predictors with outputs of 0.85, 0.87, and 0.83 respectively. The fusion uses the weighted average method to calculate the final prediction result of 0.85.

[0062] Multiple spray balance predictors are merged as independent parallel nodes to form a spray balance prediction processing layer, which processes the input desulfurization tower monitoring data sample set and spray decision sample set in parallel and generates multiple different spray balance prediction values. In other words, multiple trained spray balance predictors are used as independent nodes, and parallel computing is adopted so that each predictor processes the input data independently and finally merges the results. The output layer of the spray balance prediction processing layer is merged with the input layer of the spray balance prediction fusion device to generate a spray balance prediction branch, which is an overall module containing multiple independent processing nodes (spray balance predictors) and a fusion device, and is used to efficiently predict spray balance in practical applications. Through the parallel computing of multiple spray balance predictors, a large amount of monitoring data can be processed quickly, multiple spray balance predictors can be integrated and learned, and the results of multiple predictors can be reasonably integrated, thereby reducing the possible errors of a single prediction model and improving the stability and accuracy of the prediction.

[0063] Further, step five of this application includes:

[0064] Based on the multiple desulfurization tower monitoring data, the first desulfurization spray decision space is constrainedly optimized according to the desulfurization spray prediction channel and the desulfurization spray constraint optimization channel to establish a first desulfurization spray optimization space; according to the desulfurization spray fitness optimization channel, the first desulfurization spray optimization space is optimized for maximizing the desulfurization spray fitness to generate a first desulfurization spray optimization strategy; the first desulfurization spray optimization strategy is added to the multiple desulfurization spray optimization strategies.

[0065] Specifically, for the first desulfurization spray decision space, the monitoring data of multiple desulfurization towers and each decision plan are input into the desulfurization spray prediction channel and the desulfurization spray constraint optimization channel respectively to find the spray decision that meets the constraint conditions. Input multiple desulfurization tower monitoring data and decision plans into the desulfurization spray prediction channel, make predictions, and obtain the performance indicators of each spray decision (such as spray balance, redundancy, and energy efficiency). Input the prediction results into the desulfurization spray constraint optimization channel to determine whether each spray decision meets the predetermined constraints. Only when each indicator meets the expected interval can the decision plan be added to the optimization space. By performing optimization analysis in the first desulfurization spray decision space, a spray decision set that meets the conditions is determined, which is called the first desulfurization spray optimization space, which contains multiple spray decisions that meet the constraints.

[0066] Input the first desulfurization spray optimization space into the desulfurization spray fitness optimization channel, maximize the optimization of each spray decision, and find the best control strategy. Through the desulfurization spray fitness optimization channel, according to the previously set spray balance weight, spray margin weight and spray energy efficiency weight, the first desulfurization spray optimization space is optimized, and finally the most suitable first desulfurization spray optimization strategy is generated. The fitness of each spray decision is evaluated using the weights of the evaluation factors (such as the weights of spray balance, spray margin and spray energy efficiency), and the fitness score is obtained according to the weights of the desulfurization spray multivariate evaluation factors. Select the spray decision with the highest fitness score as the first desulfurization spray optimization strategy.

[0067] Add the first desulfurization spray optimization strategy to multiple desulfurization spray optimization strategies. Perform the above steps for other desulfurization spray decision spaces in multiple desulfurization spray decision spaces, and obtain the desulfurization spray optimization strategies corresponding to each desulfurization spray decision space, which are added to multiple desulfurization spray optimization strategies to form an overall optimization strategy, realize multi-tower coordinated control, and improve the efficiency and stability of the entire multi-tower desulfurization spray device. Through constraint and fitness optimization, the most suitable spray control strategy is finally selected to achieve the best spray effect for each desulfurization tower. The spray decision is dynamically adjusted according to the status and demand of different desulfurization towers to ensure that the multi-tower desulfurization spray device can operate efficiently under various working conditions.

[0068] Furthermore, the present application also includes the following steps:

[0069] According to the first desulfurization spray decision space, the first spray decision is randomly extracted; the first desulfurization tower monitoring data and the first spray decision are input into the desulfurization spray prediction channel to obtain the first desulfurization spray prediction result; the first desulfurization spray prediction result is input into the desulfurization spray constraint optimization channel to determine whether the first desulfurization spray prediction result satisfies the first desulfurization spray constraint in the desulfurization spray constraint optimization channel; if the first desulfurization spray prediction result satisfies the first desulfurization spray constraint, the first spray decision is added to the first desulfurization spray optimization space; if the first desulfurization spray prediction result does not satisfy the first desulfurization spray constraint, the first spray decision is eliminated; based on the first desulfurization tower monitoring data, the first desulfurization spray decision space is continued to be optimized according to the desulfurization spray prediction channel and the desulfurization spray constraint optimization channel to generate the first desulfurization spray optimization space.

[0070] Specifically, the first desulfurization spray decision space refers to the spray decisions contained in any one of the multiple desulfurization spray decision spaces. Each spray decision space is a set of spray control strategies for different desulfurization towers, including parameters such as spray liquid flow, spray angle, and spray duration that each desulfurization tower needs to adopt. A spray decision is randomly selected from the first desulfurization spray decision space as the first spray decision, and a control scheme is generated by simulating different working conditions and needs. The first desulfurization tower monitoring data and the first spray decision are input into the desulfurization spray prediction channel to simulate and predict the impact of the spray control strategy on the desulfurization tower. Based on the input monitoring data (such as gas volume, temperature, humidity, etc. in the tower) and spray decisions (such as flow, duration, etc.), the prediction channel will generate three main spray effect indicators, including spray balance prediction, spray sufficiency prediction, and spray energy efficiency prediction, which constitute the first desulfurization spray prediction result.

[0071] The first desulfurization spray prediction result is input into the desulfurization spray constraint optimization channel to determine whether the first desulfurization spray prediction result satisfies the first desulfurization spray constraint in the desulfurization spray constraint optimization channel, that is, to determine whether the spray balance prediction, spray margin prediction and spray energy efficiency prediction meet the spray balance expected interval, spray margin expected interval and spray energy efficiency expected interval corresponding to each biological desulfurization tower. If the prediction result satisfies all the set spray constraints (that is, the spray balance, spray margin and spray energy efficiency are all within the expected range), the first spray decision is added to the first desulfurization spray optimization space.

[0072] If any or all of the first desulfurization spray prediction results do not meet the first desulfurization spray constraint, the first spray decision is eliminated and a new spray decision is reselected for testing until the best spray decision that meets the constraint is found. Based on the monitoring data of the first desulfurization tower and the spray decisions in the spray decision space, all decisions in the first desulfurization spray decision space are optimized with the assistance of the desulfurization spray prediction channel and the spray constraint optimization channel to screen out the decisions that meet the above requirements and generate the first desulfurization spray optimization space, which includes all spray decisions that meet the constraints in the first desulfurization spray decision space. By gradually screening the spray decisions that meet the constraints, it is ensured that the selected spray decision meets the actual needs of the desulfurization tower, thereby improving the accuracy of the spray control, effectively optimizing the spray control decisions of each desulfurization tower, and ensuring that the spray system can achieve the optimal desulfurization effect under the premise of meeting the constraints of balance, redundancy and energy efficiency.

[0073] In summary, the intelligent control method of the multi-tower desulfurization spray device provided in this application has the following technical effects:

[0074] The multi-tower desulfurization spray device includes multiple spray pumps corresponding to multiple biological desulfurization towers; the multiple biological desulfurization towers are monitored to obtain multiple desulfurization tower monitoring data; according to the multiple desulfurization tower monitoring data, control decisions are made for the multiple spray pumps respectively, and multiple desulfurization spray decision spaces are established; a desulfurization spray multivariate evaluation factor is set, wherein the desulfurization spray multivariate evaluation factor includes spray balance, spray margin and spray energy efficiency; according to the desulfurization spray multivariate evaluation factor, a desulfurization spray multi-layer optimization channel is built; based on the multiple desulfurization tower monitoring data and according to the desulfurization spray multi-layer optimization channel, the multiple desulfurization spray decision spaces are respectively optimized at multiple levels to obtain multiple desulfurization spray optimization strategies; based on the multiple desulfurization spray optimization strategies, the multiple biological desulfurization towers are sprayed according to the multi-tower desulfurization spray device. In other words, a decision space is established for each desulfurization tower through monitoring data, multi-source evaluation factors are set, a multi-layer optimization channel is constructed, multi-level joint optimization is performed on multiple decision spaces, the operation of each spray pump is adjusted, and the optimal spraying of multiple biological desulfurization towers is achieved, thereby improving the desulfurization efficiency.

[0075] Embodiment 2: Based on the same inventive concept as the intelligent control method of the multi-tower desulfurization spray device in the aforementioned embodiment 1, the present application also provides an intelligent control system for the multi-tower desulfurization spray device, please refer to the attached Figure 2 The intelligent control system of the multi-tower desulfurization spray device includes:

[0076] A monitoring data acquisition module 11, the monitoring data acquisition module 11 is used to monitor the multiple biological desulfurization towers and obtain multiple desulfurization tower monitoring data; a decision space establishment module 12, the decision space establishment module 12 is used to make control decisions on the multiple spray pumps according to the multiple desulfurization tower monitoring data, and establish multiple desulfurization spray decision spaces; an evaluation factor setting module 13, the evaluation factor setting module 13 is used to set a desulfurization spray multivariate evaluation factor, wherein the desulfurization spray multivariate evaluation factor includes spray balance, spray sufficiency and spray energy efficiency; optimal channel building A construction module 14, the optimization channel construction module 14 is used to build a desulfurization spray multi-layer optimization channel according to the desulfurization spray multivariate evaluation factors; a joint optimization module 15, the joint optimization module 15 is used to, based on the multiple desulfurization tower monitoring data and according to the desulfurization spray multi-layer optimization channel, perform multi-level joint optimization on the multiple desulfurization spray decision spaces respectively to obtain multiple desulfurization spray optimization strategies; a strategy execution module 16, the strategy execution module 16 is used to spray the multiple biological desulfurization towers according to the multi-tower desulfurization spray device based on the multiple desulfurization spray optimization strategies.

[0077] Furthermore, the decision space establishment module 12 in the intelligent control system of the multi-tower desulfurization spray device is also used for:

[0078] Traverse the multiple biological desulfurization towers and extract the first biological desulfurization tower; extract the first desulfurization tower monitoring data corresponding to the first biological desulfurization tower based on the multiple desulfurization tower monitoring data; use the first desulfurization tower monitoring data as the spray control retrieval constraint; use the control parameters of the spray pump corresponding to the first desulfurization tower monitoring data as the spray control retrieval target; perform spray control record retrieval according to the spray control retrieval constraint and the spray control retrieval target to obtain a spray control scheme record set; perform multi-dimensional spray control interval identification according to the spray control scheme record set to generate a spray control trigger domain; use the first desulfurization tower monitoring data as the spray control target, make a control decision on the spray control trigger domain, generate a first desulfurization spray decision space, and add the first desulfurization spray decision space to the multiple desulfurization spray decision spaces.

[0079] Furthermore, the optimal channel building module 14 in the intelligent control system of the multi-tower desulfurization spray device is also used for:

[0080] Evaluation feature learning is performed according to the desulfurization spray multivariate evaluation factors to establish a desulfurization spray prediction channel; the multiple biological desulfurization towers are expected to be set according to the desulfurization spray multivariate evaluation factors to establish a desulfurization spray constraint optimization channel, wherein the desulfurization spray constraint optimization channel includes multiple desulfurization spray constraints, and each desulfurization spray constraint includes an expected spray balance interval, an expected spray margin interval, and an expected spray energy efficiency interval corresponding to each biological desulfurization tower; importance evaluation is performed according to the desulfurization spray multivariate evaluation factors to obtain the importance evaluation coefficient of each factor; weights are assigned to the desulfurization spray multivariate evaluation factors according to the importance evaluation coefficients of each factor to construct a desulfurization spray fitness optimization channel, wherein the desulfurization spray fitness optimization channel includes a spray balance weight, a spray margin weight, and a spray energy efficiency weight; the desulfurization spray prediction channel, the desulfurization spray constraint optimization channel, and the desulfurization spray fitness optimization channel are connected to generate the desulfurization spray multi-layer optimization channel.

[0081] Furthermore, the optimal channel building module 14 in the intelligent control system of the multi-tower desulfurization spray device is also used for:

[0082] According to the desulfurization spray multivariate evaluation factor, the desulfurization spray evaluation record is retrieved to obtain the desulfurization tower monitoring data sample set, the spray decision sample set, the spray balance sample set, the spray margin sample set and the spray energy efficiency sample set; based on K spray evaluation feature learning models, the desulfurization tower monitoring data sample set, the spray decision sample set and the spray balance sample set are integrated and fused to generate a spray balance prediction branch, wherein K is a positive integer greater than 1; based on the ... The tower monitoring data sample set, the spray decision sample set and the spray margin sample set are integrated and fused to generate a spray margin prediction branch; based on the K spray evaluation feature learning models, the desulfurization tower monitoring data sample set, the spray decision sample set and the spray energy efficiency sample set are integrated and fused to generate a spray energy efficiency prediction branch; the spray balance prediction branch, the spray margin prediction branch and the spray energy efficiency prediction branch are connected as parallel nodes to generate the desulfurization spray prediction channel.

[0083] Furthermore, the optimal channel building module 14 in the intelligent control system of the multi-tower desulfurization spray device is also used for:

[0084] According to the desulfurization tower monitoring data sample set, the spray decision sample set and the spray balance sample set, supervised training is performed on the K spray evaluation feature learning models to generate multiple spray balance predictors; the output data sets of the multiple spray balance predictors are used as input information, and the spray balance sample set is used as output information to train a spray balance prediction fusion device; the multiple spray balance predictors are merged as parallel independent nodes to generate a spray balance prediction processing layer; the spray balance prediction processing layer is merged with the input layer of the spray balance prediction fusion device to generate the spray balance prediction branch.

[0085] Furthermore, the joint optimization module 15 in the intelligent control system of the multi-tower desulfurization spray device is also used for:

[0086] Based on the multiple desulfurization tower monitoring data, the first desulfurization spray decision space is constrainedly optimized according to the desulfurization spray prediction channel and the desulfurization spray constraint optimization channel to establish a first desulfurization spray optimization space; according to the desulfurization spray fitness optimization channel, the first desulfurization spray optimization space is optimized for maximizing the desulfurization spray fitness to generate a first desulfurization spray optimization strategy; the first desulfurization spray optimization strategy is added to the multiple desulfurization spray optimization strategies.

[0087] Furthermore, the joint optimization module 15 in the intelligent control system of the multi-tower desulfurization spray device is also used for:

[0088] According to the first desulfurization spray decision space, the first spray decision is randomly extracted; the first desulfurization tower monitoring data and the first spray decision are input into the desulfurization spray prediction channel to obtain the first desulfurization spray prediction result; the first desulfurization spray prediction result is input into the desulfurization spray constraint optimization channel to determine whether the first desulfurization spray prediction result satisfies the first desulfurization spray constraint in the desulfurization spray constraint optimization channel; if the first desulfurization spray prediction result satisfies the first desulfurization spray constraint, the first spray decision is added to the first desulfurization spray optimization space; if the first desulfurization spray prediction result does not satisfy the first desulfurization spray constraint, the first spray decision is eliminated; based on the first desulfurization tower monitoring data, the first desulfurization spray decision space is continued to be optimized according to the desulfurization spray prediction channel and the desulfurization spray constraint optimization channel to generate the first desulfurization spray optimization space.

[0089] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The intelligent control method and specific examples of the multi-tower desulfurization spray device in Example 1 are also applicable to the intelligent control system of the multi-tower desulfurization spray device in this embodiment. Through the detailed description of the intelligent control method of the multi-tower desulfurization spray device, those skilled in the art can clearly know the intelligent control system of the multi-tower desulfurization spray device in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0090] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0091] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. An intelligent control method for a multi-tower desulfurization spray device, characterized in that: The multi-tower desulfurization spray device includes a plurality of spray pumps corresponding to a plurality of biological desulfurization towers, and the intelligent control method of the multi-tower desulfurization spray device includes: Monitoring the multiple biological desulfurization towers to obtain multiple desulfurization tower monitoring data; According to the monitoring data of the multiple desulfurization towers, control decisions are made on the multiple spray pumps respectively to establish multiple desulfurization spray decision spaces; Setting a desulfurization spray multivariate evaluation factor, wherein the desulfurization spray multivariate evaluation factor includes spray balance, spray margin and spray energy efficiency; According to the desulfurization spray multivariate evaluation factors, a desulfurization spray multi-layer optimization channel is established; Based on the monitoring data of the multiple desulfurization towers, and according to the desulfurization spray multi-layer optimization channel, the multiple desulfurization spray decision spaces are respectively optimized at multiple levels to obtain multiple desulfurization spray optimization strategies; Based on the multiple desulfurization spray optimization strategies, spraying the multiple biological desulfurization towers according to the multi-tower desulfurization spray device; According to the desulfurization spray multi-evaluation factors, a desulfurization spray multi-layer optimization channel is established, including: Conducting evaluation feature learning based on the desulfurization spray multivariate evaluation factors to establish a desulfurization spray prediction channel; According to the desulfurization spray multivariate evaluation factor, the multiple biological desulfurization towers are expected to be set, and a desulfurization spray constraint optimization channel is established, wherein the desulfurization spray constraint optimization channel includes multiple desulfurization spray constraints, and each desulfurization spray constraint includes an expected interval of spray balance, an expected interval of spray sufficiency, and an expected interval of spray energy efficiency corresponding to each biological desulfurization tower; Performing importance evaluation according to the desulfurization spray multivariate evaluation factors to obtain importance evaluation coefficients of each factor; According to the importance evaluation coefficient of each factor, the desulfurization spray multivariate evaluation factor is weighted and a desulfurization spray fitness optimization channel is constructed, wherein the desulfurization spray fitness optimization channel includes a spray balance weight, a spray sufficiency weight and a spray energy efficiency weight; The desulfurization spray prediction channel, the desulfurization spray constraint optimization channel and the desulfurization spray fitness optimization channel are connected to generate the desulfurization spray multi-layer optimization channel.

2. The intelligent control method for a multi-tower desulfurization spray device according to claim 1, characterized in that: According to the monitoring data of the multiple desulfurization towers, control decisions are made on the multiple spray pumps respectively, and multiple desulfurization spray decision spaces are established, including: Traversing the plurality of biological desulfurization towers, extracting a first biological desulfurization tower; Extracting the first desulfurization tower monitoring data corresponding to the first biological desulfurization tower according to the plurality of desulfurization tower monitoring data; Using the first desulfurization tower monitoring data as a spray control retrieval constraint; Taking the control parameter of the spray pump corresponding to the monitoring data of the first desulfurization tower as the spray control retrieval target; Perform spray control record retrieval according to the spray control retrieval constraint and the spray control retrieval target to obtain a spray control scheme record set; Perform multi-dimensional spray control interval identification according to the spray control scheme record set to generate a spray control trigger domain; Taking the first desulfurization tower monitoring data as the spray control target, a control decision is made on the spray control trigger domain to generate a first desulfurization spray decision space, and the first desulfurization spray decision space is added to the multiple desulfurization spray decision spaces.

3. The intelligent control method for a multi-tower desulfurization spray device according to claim 1, characterized in that: Evaluation feature learning is performed according to the desulfurization spray multivariate evaluation factor to establish a desulfurization spray prediction channel, including: Retrieve desulfurization spray evaluation records according to the desulfurization spray multivariate evaluation factors to obtain a desulfurization tower monitoring data sample set, a spray decision sample set, a spray balance sample set, a spray adequacy sample set, and a spray energy efficiency sample set; Based on K spray evaluation feature learning models, the desulfurization tower monitoring data sample set, the spray decision sample set and the spray balance sample set are integrated and fused to generate a spray balance prediction branch, wherein K is a positive integer greater than 1; Based on the K spray evaluation feature learning models, the desulfurization tower monitoring data sample set, the spray decision sample set and the spray margin sample set are integrated and fused to learn, and a spray margin prediction branch is generated; Based on the K spray evaluation feature learning models, the desulfurization tower monitoring data sample set, the spray decision sample set and the spray energy efficiency sample set are integrated and fused to learn, and a spray energy efficiency prediction branch is generated; The spray balance prediction branch, the spray sufficiency prediction branch and the spray energy efficiency prediction branch are connected as parallel nodes to generate the desulfurization spray prediction channel.

4. The intelligent control method for a multi-tower desulfurization spraying device according to claim 3, characterized in that: Based on K spray evaluation feature learning models, the desulfurization tower monitoring data sample set, the spray decision sample set and the spray balance sample set are integrated and fused to generate a spray balance prediction branch, including: According to the desulfurization tower monitoring data sample set, the spray decision sample set and the spray balance sample set, supervised training is performed on the K spray evaluation feature learning models to generate multiple spray balance predictors; Using the output data sets of the plurality of spray balance predictors as input information and the spray balance sample set as output information, training a spray balance prediction fusion device; The plurality of spray balance predictors are merged as parallel independent nodes to generate a spray balance prediction processing layer; The spray balance prediction processing layer is merged with the input layer of the spray balance prediction fusion device to generate the spray balance prediction branch.

5. The intelligent control method for a multi-tower desulfurization spraying device according to claim 2, characterized in that: Based on the monitoring data of the multiple desulfurization towers and according to the desulfurization spray multi-layer optimization channel, the multiple desulfurization spray decision spaces are respectively optimized at multiple levels to obtain multiple desulfurization spray optimization strategies, including: Based on the monitoring data of the multiple desulfurization towers, constrained optimization is performed on the first desulfurization spray decision space according to the desulfurization spray prediction channel and the desulfurization spray constraint optimization channel to establish a first desulfurization spray optimization space; According to the desulfurization spray fitness optimization channel, the first desulfurization spray optimization space is optimized for maximizing the desulfurization spray fitness to generate a first desulfurization spray optimization strategy; The first desulfurization spray optimization strategy is added to the multiple desulfurization spray optimization strategies.

6. The intelligent control method for a multi-tower desulfurization spraying device according to claim 5, characterized in that: Based on the plurality of desulfurization tower monitoring data, the first desulfurization spray decision space is constrainedly optimized according to the desulfurization spray prediction channel and the desulfurization spray constraint optimization channel to establish the first desulfurization spray optimization space, including: According to the first desulfurization spray decision space, randomly extracting a first spray decision; Inputting the first desulfurization tower monitoring data and the first spray decision into the desulfurization spray prediction channel to obtain a first desulfurization spray prediction result; Inputting the first desulfurization spray prediction result into the desulfurization spray constraint optimization channel to determine whether the first desulfurization spray prediction result satisfies the first desulfurization spray constraint in the desulfurization spray constraint optimization channel; If the first desulfurization spray prediction result satisfies the first desulfurization spray constraint, adding the first spray decision to the first desulfurization spray optimization space; If the first desulfurization spray prediction result does not meet the first desulfurization spray constraint, the first spray decision is eliminated; Based on the first desulfurization tower monitoring data, the first desulfurization spray decision space is continuously optimized and analyzed according to the desulfurization spray prediction channel and the desulfurization spray constraint optimization channel to generate the first desulfurization spray optimization space.

7. The intelligent control system of the multi-tower desulfurization spray device is characterized by: Steps for implementing the intelligent control method of the multi-tower desulfurization spray device according to any one of claims 1 to 6, wherein the intelligent control system of the multi-tower desulfurization spray device comprises: A monitoring data acquisition module, wherein the monitoring data acquisition module is used to monitor the multiple biological desulfurization towers and obtain multiple desulfurization tower monitoring data; A decision space establishment module, the decision space establishment module is used to make control decisions on the multiple spray pumps respectively according to the multiple desulfurization tower monitoring data, and establish multiple desulfurization spray decision spaces; An evaluation factor setting module, wherein the evaluation factor setting module is used to set a desulfurization spray multivariate evaluation factor, wherein the desulfurization spray multivariate evaluation factor includes spray balance, spray margin and spray energy efficiency; An optimization channel building module, wherein the optimization channel building module is used to build a desulfurization spray multi-layer optimization channel according to the desulfurization spray multi-evaluation factor; A joint optimization module, wherein the joint optimization module is used to perform multi-level joint optimization on the multiple desulfurization spray decision spaces based on the multiple desulfurization tower monitoring data and the desulfurization spray multi-layer optimization channel to obtain multiple desulfurization spray optimization strategies; A strategy execution module is used to spray the multiple biological desulfurization towers according to the multi-tower desulfurization spray device based on the multiple desulfurization spray optimization strategies.

Images