Control method for main reaction device of calcium hypochlorite based on reaction variable monitoring

Abstract

The invention relates to the technical field of chemical production, in particular to a control method for a main reaction device of calcium hypochlorite based on reaction variable monitoring. The future particle agglomeration state is predicted through a particle agglomeration time sequence state prediction model, a state hot spot chromaticity diagram of actual particle agglomeration is interpolated according to prediction result space chromaticity, and the stirring rate and an air escape valve of a bleaching powder main reaction device are controlled based on the state hot spot chromaticity diagram; and constructing a gas generation quantitative hash table when the side reaction occurs, performing hash accumulation on the gas generated when the side reaction occurs in the main reaction component by using the gas generation quantitative hash table, and analyzing the reaction temperature of the control device according to the hash accumulation result. According to the invention, the main reaction device can be subjected to stirring, gas leakage and accurate control of reaction temperature according to particle agglomeration and gas dissipation phenomena in the production process of the high test bleaching powder, so that the production safety and efficiency of the main reaction device of the high test bleaching powder are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical production, and particularly to a control method for the main reaction device of bleaching powder concentrate based on reaction variable monitoring. Background Art

[0002] Bleaching powder concentrate is a chemical widely used in water treatment, disinfection, bleaching and other fields. The operating state of the main reaction device in its production process directly affects product quality, production efficiency and energy consumption. In recent years, with the development of intelligent manufacturing and automation technologies, by real-time monitoring key variables (such as pH value, redox potential, and concentration, etc.) during the reaction process and making intelligent adjustments, the stability and automation level of the production process can be improved. However, the existing main reaction devices of bleaching powder concentrate usually adopt a method of setting fixed process parameters for control, such as temperature, pressure, feeding rate, etc. Due to possible influences of factors such as raw material purity, environmental conditions, and equipment aging during the reaction process, the process parameters may fluctuate, thereby affecting the stability of product quality. Moreover, the existing devices pay too much attention to basic process parameters such as temperature and pressure during the preparation process, and less monitor key chemical indicators, particle agglomeration, and side reaction decomposition and other phenomena during the reaction process, resulting in the inability to detect abnormalities in a timely manner. At the same time, the existing equipment still relies on manual sampling analysis to adjust process parameters, which has a certain lag and cannot achieve fine control. Summary of the Invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a control method for the main reaction device of bleaching powder concentrate based on reaction variable monitoring.

[0004] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0005] The first aspect of the present invention provides a control method for the main reaction device of bleaching powder concentrate based on reaction variable monitoring, including the following steps:

[0006] Obtain the reaction variable monitoring data of bleaching powder concentrate, and perform state random deduction on the reaction variable monitoring data through constructing a standard reaction stochastic evolution equation of bleaching powder concentrate to obtain the simulated reaction state of bleaching powder concentrate in the device;

[0007] According to the adjacency topology of the simulated reaction state on different established monitoring nodes, calculate the state transition value of particle agglomeration occurring in the bleaching powder concentrate reaction to construct a particle agglomeration time series state prediction model of the bleaching powder concentrate reaction;

[0008] Predict the future particle agglomeration state through the particle agglomeration time series state prediction model, interpolate the actual particle agglomeration state hot chromaticity map according to the prediction result hot chromaticity, and control the stirring rate and air release valve of the main reaction device of bleaching powder concentrate based on the state hot chromaticity map;

[0009] Construct a quantitative hash table for gas generation during side reactions. Use the quantitative hash table for gas generation to hash and accumulate the gases produced when side reactions occur in the main reaction components, and analyze the reaction temperature of the control device based on the hash accumulation result.

[0010] More specifically, for obtaining the reaction variable monitoring data of bleaching powder concentrate, perform state random deduction on the reaction variable monitoring data by constructing a standard reaction stochastic evolution equation of bleaching powder concentrate to obtain the simulated reaction state of bleaching powder concentrate in the device, which specifically includes the following steps:

[0011] Obtain the standard reaction process of bleaching powder concentrate, and extract the main reaction components of bleaching powder concentrate and the reference reaction concentration and accurate rate constant of the main reaction components when producing qualified bleaching powder concentrate through the standard reaction process;

[0012] Search and obtain the reaction kinetic formula of the main reaction components in the big data network based on the standard reaction process, and perform dynamic calculation on the reference reaction concentration and accurate rate constant through the reaction kinetic formula to obtain the standard reaction tendency function;

[0013] Obtain the reaction variable monitoring data and reaction condition process of the main reaction device of bleaching powder concentrate, introduce the molecular stochastic evolution model, and perform evolution description on the reaction condition process in the molecular stochastic evolution model based on the standard reaction process to output the standard reaction stochastic evolution equation of bleaching powder concentrate;

[0014] Solve the standard reaction stochastic evolution equation through the standard reaction tendency function to obtain the Poisson trigger time step of the next main reaction. Based on the reaction variable monitoring data, perform cumulative calculation on the reaction rate distribution of the main reaction components. During the cumulative process, advance the time sequence of the standard reaction stochastic evolution equation according to the Poisson trigger time step of the next main reaction;

[0015] After advancing to reach the reaction termination step length, output the evolution reactions of the main reaction components under a series of cumulative rate distributions, and update the state of the standard reaction stochastic evolution equation according to the evolution reactions to obtain the simulated reaction state of bleaching powder concentrate in the device.

[0016] More specifically, for calculating the state transition value of particle agglomeration when the bleaching powder concentrate reaction occurs according to the simulated reaction state adjacency topology on different established monitoring nodes to construct a particle agglomeration time sequence state prediction model of the bleaching powder concentrate reaction, which specifically includes the following steps:

[0017] Obtain the preset monitoring strategy for monitoring reaction variables of the main reaction device of bleaching powder concentrate, and extract the continuous established monitoring nodes of the main reaction device of bleaching powder concentrate through the preset monitoring strategy;

[0018] Obtain the simulated reaction state of the main reaction device of bleaching powder concentrate at each established monitoring node, which is defined as the current simulated reaction state. Allocate and calculate the state boundary weight for the transfer of bleaching powder concentrate from the current simulated reaction state at the established monitoring node to the next established monitoring node according to the reaction variable monitoring data;

[0019] Using the current simulated reaction state as the state adjacent loop node, connect each state adjacent loop node based on the state boundary weight to construct the state transfer adjacent graph of the current simulated reaction state, and strip out the state topological layout of the bleaching powder concentrate reaction through the state transfer adjacent graph;

[0020] Based on the big data network, obtain the approaching reaction conditions for the particle agglomeration phenomenon of bleaching powder concentrate. Construct a reaction state evaluation system for the particle agglomeration in the bleaching powder concentrate reaction according to the approaching reaction conditions. Introduce the Bellman equation, and evaluate and update the value of each transfer state in the state topological layout in the Bellman equation based on the reaction state evaluation system to obtain the value function of the transfer state;

[0021] Extract the maximum critical criterion of the reaction state evaluation system, preset the value function threshold based on the maximum critical criterion. If the value function is greater than the value function threshold, repeat the above evaluation and update iteration steps of the value function of the transfer state until it is less than the value function threshold, obtain the value function matrix, and train and construct a particle agglomeration time series state prediction model for the bleaching powder concentrate reaction based on the value function matrix.

[0022] More specifically, predicting the future particle agglomeration state through the particle agglomeration time series state prediction model, interpolating the actual particle agglomeration state hot spot chromaticity map according to the prediction result space chromaticity, and controlling the stirring rate and air release valve of the main reaction device of bleaching powder concentrate based on the state hot spot chromaticity map, specifically including the following steps:

[0023] Obtain the production demand of bleaching powder concentrate, preset the future reaction time series according to the production demand, predict the particle agglomeration of the main reaction components of bleaching powder concentrate on the future reaction time series through the particle agglomeration time series state prediction model, and output the particle agglomeration state segment of the main reaction components on the future reaction time series, marked as the future particle agglomeration state slice;

[0024] Obtain the specification variation index of different preset particle agglomeration particle sizes when the main reaction components produce bleaching powder concentrate through big data, and assign the corresponding hot spot chromaticity components to each preset particle agglomeration particle size based on the specification variation index;

[0025] Introduce the Manhattan distance method to calculate the spatial correlation of the actual particle agglomeration particle size in each future particle agglomeration state slice, generate the covariance matrix of the future particle agglomeration state slice, calculate the particle agglomeration particle size pattern in the covariance matrix based on the hot spot chromaticity components, and obtain the Kriging equations of the particle agglomeration particle size pattern;

[0026] Solve the Kriging equations simultaneously to obtain the component interpolation weights of different actual particle agglomeration diameters in each future particle agglomeration state slice. Interpolate the hot spot chromaticity components corresponding to the hot spots to the actual particle agglomeration diameters in each particle agglomeration diameter pattern according to the component interpolation weights, and generate the state hot spot chromaticity map of the actual particle agglomeration;

[0027] Obtain the normal particle agglomeration state of bleaching powder concentrate through production requirements, preset the normal hot spot color gamut interval according to the normal particle agglomeration state. If the current state hot spot chromaticity shown in the state hot spot chromaticity map of the actual particle agglomeration cannot be queried within the normal hot spot chromaticity interval, control the stirring rate of the main reaction device of bleaching powder concentrate and open the air release valve.

[0028] More specifically, construct a gas generation quantization hash table when side reactions occur, use the gas generation quantization hash table to perform hash accumulation on the gases generated when side reactions occur in the main reaction components, and analyze and control the reaction temperature of the device according to the hash accumulation result, specifically including the following steps:

[0029] Construct an initial gas quantization hash table, perform hash balancing quantization calculation on the reaction variable monitoring data based on the side chemical reaction balancing formula to obtain gas generation molar hash key-value pairs, and analyze and store the gas generation molar hash key-value pairs in the initial gas quantization hash table to obtain the gas generation quantization hash table when side reactions occur;

[0030] When the air release valve of the main reaction device of bleaching powder concentrate is in the open state, split the reaction variable monitoring data into N data blocks based on consecutive established monitoring nodes of the main reaction device of bleaching powder concentrate, and construct a cumulative leaf node according to each data block;

[0031] Perform hash query on each cumulative leaf node through the gas generation quantization hash table to obtain the gas generation quantization hash value of each cumulative leaf node;

[0032] Starting from a certain cumulative leaf node, combine the adjacent two gas generation quantization hash values to generate a new hash combination of this cumulative leaf node, and use the gas generation quantization hash table to calculate the hash value of the new hash combination again to form a new gas generation quantization hash value;

[0033] Use the new gas generation quantization hash value as the cumulative leaf node of the upper layer, and continuously repeat the above steps of combining adjacent gas generation quantization hash values and recalculating the hash value until a unique root hash value is generated, to obtain the gas cumulative root hash value when side reactions occur;

[0034] Identify the gas cumulative root hash value through the chemical reaction knowledge graph to obtain the temperature gradient that causes the generation of the gas cumulative root hash value, and construct it as the first temperature gradient curve;

[0035] Obtain the current temperature gradient of the calcium hypochlorite main reaction device, construct it as the second temperature gradient curve, calculate the slope difference between the first temperature gradient curve and the second temperature gradient curve to obtain the slope difference value, and control the internal reaction temperature of the calcium hypochlorite main reaction device according to the slope difference value.

[0036] More specifically, the construction of the gas quantification initial hash table is based on the side chemical reaction balancing formula to perform hash balancing quantification calculation on the reaction variable monitoring data, obtain the gas generation molar hash key-value pair, analyze and store the gas generation molar hash key-value pair in the gas quantification initial hash table, and obtain the gas generation quantification hash table when a side reaction occurs. The specific steps are as follows:

[0037] Obtain the chemical reaction knowledge graph based on the big data network, identify the main reaction components through the chemical reaction knowledge graph, and output the reaction characteristics when the main reaction components are incompletely dissolved;

[0038] Introduce the gas side reaction molar law, use the reaction characteristics as the indexing rule, and construct the storage key table structure of the gas side reaction molar law based on the indexing rule to generate the gas quantification initial hash table;

[0039] Obtain the side chemical reaction balancing formula when the main reaction components are incompletely dissolved, introduce the encryption hash algorithm, and use the encryption hash algorithm to perform balancing quantification calculation on the reaction variable monitoring data in the side reaction chemical formula to obtain the gas generation molar hash key-value pair when the main reaction components are incompletely dissolved under different reaction variable monitoring data conditions;

[0040] If the index position of the gas quantification initial hash table for the gas generation molar hash key-value pair is empty, directly store the gas generation molar hash key to the current index position to obtain a type of gas hash quantification;

[0041] If the gas quantification initial hash table maps the gas generation molar hash key-value pair to the same index, perform quadratic probing to find the new storage position of the gas generation molar hash key to obtain a second type of gas hash quantification;

[0042] Perform dynamic update and adjustment on the gas quantification initial hash table based on the first type of gas hash quantification and the second type of gas hash quantification to obtain the gas generation quantification hash table when a side reaction occurs.

[0043] In the second aspect of the present invention, a control system for the main reaction device of bleaching powder essence based on reaction variable monitoring is provided. The control system for the main reaction device of bleaching powder essence includes a memory and a processor. A control method program for the main reaction device of bleaching powder essence based on reaction variable monitoring is stored in the memory. When the control method program for the main reaction device of bleaching powder essence is executed by the processor, the steps of any of the control methods for the main reaction device of bleaching powder essence are implemented.

[0044] The present invention solves the technical defects existing in the background art. The beneficial technical effects of the present invention are as follows:

[0045] Obtain the reaction variable monitoring data of bleaching powder essence, perform state random deduction on the reaction variable monitoring data through constructing a standard reaction stochastic evolution equation of bleaching powder essence to obtain the simulated reaction state of bleaching powder essence in the device; calculate the state transition value of particle agglomeration in the bleaching powder essence reaction according to the adjacency topology of the simulated reaction state on different established monitoring nodes to construct a particle agglomeration time series state prediction model for the bleaching powder essence reaction; predict the future particle agglomeration state through the particle agglomeration time series state prediction model, interpolate the actual particle agglomeration state hot spot chromaticity diagram according to the prediction result space chromaticity, and control the stirring rate and air release valve of the main reaction device of bleaching powder essence based on the state hot spot chromaticity diagram; construct a gas generation quantization hash table when side reactions occur, use the gas generation quantization hash table to hash and accumulate the gases generated when side reactions occur in the main reaction components, and analyze and control the reaction temperature of the device according to the hash accumulation result. The present invention can accurately control the stirring, air release, and reaction temperature of the main reaction device according to the particle agglomeration and gas escape phenomena occurring in the production process of bleaching powder essence, so that the production reaction of the main reaction device of bleaching powder essence is more stable and efficient, and the reaction safety factor is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0047] Figure 1 Shows the first method flow chart of a control method for the main reaction device of bleaching powder essence based on reaction variable monitoring;

[0048] Figure 2 Shows the second method flow chart of a control method for the main reaction device of bleaching powder essence based on reaction variable monitoring;

[0049] Figure 3Shows a system framework diagram of a calcium hypochlorite main reaction device control system based on reaction variable monitoring. Detailed implementation manners

[0050] In order to more clearly understand the above objects, features and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation manners. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.

[0051] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0052] The first aspect of the present invention provides a control method for a calcium hypochlorite main reaction device based on reaction variable monitoring, as Figure 1 shown, including the following steps:

[0053] S102: Obtain the reaction variable monitoring data of calcium hypochlorite, perform state random deduction on the reaction variable monitoring data by constructing a standard reaction stochastic evolution equation of calcium hypochlorite, and obtain the simulated reaction state of calcium hypochlorite in the device;

[0054] S104: Calculate the state transition value of particle agglomeration in the calcium hypochlorite reaction according to the simulated reaction state adjacency topology on different established monitoring nodes, so as to construct a particle agglomeration time series state prediction model for the calcium hypochlorite reaction;

[0055] S106: Predict the future particle agglomeration state through the particle agglomeration time series state prediction model, interpolate the actual particle agglomeration state hot chromaticity map according to the prediction result hot chromaticity, and control the stirring rate and air release valve of the calcium hypochlorite main reaction device based on the state hot chromaticity map;

[0056] S108: Construct a gas generation quantization hash table when side reactions occur, use the gas generation quantization hash table to hash accumulate the gas generated when side reactions occur in the main reaction components, and analyze and control the reaction temperature of the device according to the hash accumulation result.

[0057] More specifically, the obtaining of the reaction variable monitoring data of calcium hypochlorite, performing state random deduction on the reaction variable monitoring data by constructing a standard reaction stochastic evolution equation of calcium hypochlorite, and obtaining the simulated reaction state of calcium hypochlorite in the device specifically includes the following steps:

[0058] Obtain the standard reaction process of calcium hypochlorite, and extract the main reaction components of calcium hypochlorite and the reference reaction concentration and accurate rate constant of the main reaction components when qualified calcium hypochlorite is produced through the standard reaction process;

[0059] Search for the reaction kinetic formula of the main reaction components in the big data network based on the standard reaction process, and perform dynamic calculations on the reference reaction concentration and the accurate rate constant through the reaction kinetic formula to obtain the standard reaction tendency function;

[0060] Obtain the reaction variable monitoring data and the reaction condition process of the main reaction device of bleaching powder concentrate, introduce the molecular stochastic evolution model, and perform evolutionary description on the reaction condition process in the molecular stochastic evolution model based on the standard reaction process to output the standard reaction stochastic evolution equation of bleaching powder concentrate;

[0061] Solve the standard reaction stochastic evolution equation through the standard reaction tendency function to obtain the Poisson trigger time step of the next main reaction. Based on the reaction variable monitoring data, perform cumulative calculation on the reaction rate distribution of the main reaction components. During the cumulative process, advance the time sequence of the standard reaction stochastic evolution equation according to the Poisson trigger time step of the next main reaction;

[0062] After advancing to reach the reaction termination step length, output the evolutionary reactions of the main reaction components under a series of cumulative rate distributions, and update the state of the standard reaction stochastic evolution equation according to the evolutionary reactions to obtain the simulated reaction state of bleaching powder concentrate in the device.

[0063] It should be noted that the main reaction components include lime milk (Ca(OH)2) and chlorine gas (Cl2). The reaction variable monitoring data includes concentration data and pH data. The production of bleaching powder concentrate is mainly achieved by the reaction of lime milk with chlorine gas to produce calcium hypochlorite, calcium chloride, and water, and calcium hypochlorite is the main source of bleaching powder concentrate. Therefore, the control requirements for the main reaction device are extremely strict. However, if the main reaction device of bleaching powder concentrate is not properly controlled, since lime milk is an aqueous suspension of calcium hydroxide, it may cause the chlorine gas to not react fully, affecting the yield of calcium hypochlorite. Therefore, it is particularly important to control the operation of the device in real time based on the reaction state of the main reaction components of bleaching powder concentrate. In this regard, this method first obtains the reference reaction concentration and accurate rate constant of the main reaction components when producing qualified bleaching powder concentrate according to the standard reaction process of bleaching powder concentrate, and calculates the two using the reaction kinetics formula, thereby obtaining the standard reaction tendency function. This standard reaction tendency function represents the possibility of the main reaction components undergoing the reference reaction concentration and accurate rate constant reaction under the current state, determines the time scale of the reaction, is the random evolution basis of the standard reaction state of the main reaction components, ensures the accuracy of subsequent random number sampling, and improves the deduction accuracy of the standard reaction random evolution of subsequent bleaching powder concentrate. Then, based on the standard reaction process, an evolutionary description of the reaction working condition process of the main reaction device of bleaching powder concentrate is performed in the molecular random evolution model. Here, the molecular random evolution model is a unique evolution model applied to chemical molecular reactions, making the random description of the reaction state more conform to the reaction logic of chemical molecular formulas. Finally, a system evolution equation for the deduction of the standard random state of the main reaction components by the main reaction device of bleaching powder concentrate is constructed, which can perform an import description of the reaction variable monitoring data, thereby quickly simulating the real-time state of the main reaction components when generating different reaction control monitoring data.

[0064] It should be noted that the standard reaction stochastic evolution equation further solved through the standard reaction tendency function can obtain the Poisson trigger time step of the next main reaction. It is worth mentioning that due to the uncertainty of the bleaching powder concentrate reaction, this time step follows an exponential distribution, which conforms to the characteristics of the Poisson process. The shorter the time interval between reactions means the faster the system changes. By controlling the frequencies of different reactions through the exponential distribution, the system exhibits real fluctuation behavior in time, which enables the subsequent monitoring data of the evolution reaction variables of the system to more closely follow the probability law of chemical reaction kinetics, ensuring that the simulation of the reaction state is more in line with theory and greatly improving the stochastic credibility of the simulated reaction state. Then, based on the monitoring data of the reaction variables, cumulative calculation is performed on the reaction rate distribution of the main reaction components. This means that the cumulative reaction conforms to the main reaction components generating the monitoring data of the reaction variables by following the reference reaction concentration and accurate rate constant, that is, it is equivalent to randomly selecting a most reasonable reaction according to the weight among all reactions, making the probabilities of different reactions correctly reflect their reference reaction concentration and accurate rate constant, conforming to the behavior of the real chemical system. Thus, through random selection, the linear stochastic evolution of the system for the actual state is ensured, rather than a fixed deterministic trajectory, ensuring the authenticity of the simulation of the actual reaction state. During the cumulative process, advancing the time sequence of the standard reaction stochastic evolution equation according to the Poisson trigger time step of the next main reaction can enable the system state to simulate the random fluctuations of the molecular number with time evolution, improving the time sequence of the state simulation to maintain the measurement of the reaction variables. Finally, the evolution reaction deduced by the system can further reveal the simulated reaction state of the main reaction components of the bleaching powder concentrate in the device. Through this method, a stochastic evolution system of the reaction state that conforms to the standard production of bleaching powder concentrate can be constructed to deduce the obtained monitoring data of the reaction variables, which is more accurate and efficient compared to relying on manual experience to analyze the monitoring data of the reaction variables to infer the reaction state, saves manpower output, improves the real-time intelligence accuracy of the reaction state inside the device, and provides an analysis basis for the accurate control of subsequent devices.

[0065] More specifically, calculating the state transition value of the particle agglomeration in the bleaching powder concentrate reaction according to the simulated reaction state adjacent topology on different established monitoring nodes to construct a prediction model for the sequential state of the particle agglomeration in the bleaching powder concentrate reaction, as Figure 2 shown, specifically including the following steps:

[0066] S202: Obtain the preset monitoring strategy for monitoring the reaction variables of the main reaction device of the bleaching powder concentrate, and extract the continuous established monitoring nodes of the main reaction device of the bleaching powder concentrate through the preset monitoring strategy;

[0067] S204: Obtain the simulated reaction states of the main reaction device of bleaching powder at each established monitoring node, defined as the current simulated reaction states, and allocate and calculate the state boundary weights for the transfer of bleaching powder from the current simulated reaction states at the established monitoring node to the next established monitoring node according to the reaction variable monitoring data;

[0068] S206: Take the current simulated reaction states as the state adjacent loop nodes, construct the state transfer adjacency graph of the current simulated reaction states by connecting each state adjacent loop node based on the state boundary weights, and strip out the state topological layout of the bleaching powder reaction through the state transfer adjacency graph;

[0069] S208: Obtain the approaching reaction conditions of the bleaching powder particle agglomeration phenomenon based on the big data network, construct a reaction state evaluation system for the occurrence of particle agglomeration in the bleaching powder reaction according to the approaching reaction conditions, introduce the Bellman equation, and evaluate and update the value of each transfer state in the state topological layout in the Bellman equation based on the reaction state evaluation system to obtain the value function of the transfer state;

[0070] S210: Extract the maximum critical criterion of the reaction state evaluation system, preset the value function threshold based on the maximum critical criterion. If the value function is greater than the value function threshold, repeat the above evaluation and update iteration steps of the value function of the transfer state until it is less than the value function threshold, obtain the value function matrix, and train and construct a particle agglomeration time series state prediction model for the bleaching powder reaction based on the value function matrix.

[0071] It should be noted that in the process of producing calcium hypochlorite, lime milk (Ca(OH)2) and chlorine gas (Cl2) are usually used for reaction. However, if the reaction between the two is incomplete, unreacted chlorine gas may accumulate, resulting in a synchronous increase in temperature and pressure, increasing the risk coefficient of the calcium hypochlorite production reaction. Under high-temperature conditions, calcium hypochlorite will further agglomerate into particles. Thus, the agglomeration state of the calcium hypochlorite particles produced by the reaction is one of the important bases for judging whether the internal environment control of the main reaction device of calcium hypochlorite is reasonable. Therefore, this method takes the simulated reaction states at different established monitoring nodes following the preset monitoring strategy of the main reaction device of calcium hypochlorite as the sequential input, and establishes a sequential transfer that describes the state of the main reaction components as the monitoring data of the reaction variables continuously changes; among them, since there may be gradient differences in the monitoring data of the reaction variables presented at consecutive established monitoring nodes, and the gradient differences may cause internal chemical changes in the main reaction components, thus changing the preparation state of calcium hypochlorite. Therefore, this method calculates the state boundary weight for the transfer of calcium hypochlorite from the current simulated reaction state at the established monitoring node to the next established monitoring node based on the monitoring data of the reaction variables, and constructs a state transfer adjacency graph of the current simulated reaction state by connecting each state adjacent loop node based on the state boundary weight. This state transfer adjacency graph expresses the transition of state generation, and the stripped state topological layout interprets the continuity and architecture of the sequential state changes, making the state detail changes more significant under the fixed monitoring data of the reaction variables, and improving the accuracy of the training basis for the subsequent prediction model construction.

[0072] It should be noted that the approaching reaction condition for the calcium hypochlorite particle agglomeration phenomenon refers to a critical threshold for the internal tendency to produce particle agglomeration during the dissolution of the main reaction components of calcium hypochlorite under the influence of certain factors. It is the premise basis for responding to whether the main reaction components produce particle agglomeration. Therefore, this method constructs a reaction state evaluation system for the calcium hypochlorite reaction to produce particle agglomeration based on the approaching reaction condition. For each transfer state in the state topological layout, this method uses this reaction state evaluation system to evaluate and update in the Bellman equation to generate the value of each transfer state. This value reflects the expected return of the subsequent possible state inferred from the action selection of the current transfer state. Thus, it can be known whether the transfer state under the fixed monitoring data of the reaction variables may lead to the occurrence of particle agglomeration in the next state. Each evaluation and update can improve the estimation of the transfer state, making the value function gradually approach the true value of the particle agglomeration state, so as to ensure that the inference of whether the future calcium hypochlorite reaction state will produce particle agglomeration is more accurate, and improving the prediction performance of the particle agglomeration sequential state prediction model. Through this method, the sequential transfer of the simulated reaction states at different monitoring time sequences can be analyzed and used as the state transition training data for whether the calcium hypochlorite particle agglomeration phenomenon occurs to construct a prediction model, so as to predict the actual state of the main reaction device for producing calcium hypochlorite, effectively improving the accuracy, real-time performance and rationality of the subsequent device control.

[0073] More specifically, the future particle agglomeration state is predicted by the particle agglomeration time series state prediction model, and the state hot spot chromaticity map of the actual particle agglomeration is interpolated according to the predicted result space chromaticity. Based on the state hot spot chromaticity map, the stirring rate and the air release valve of the calcium hypochlorite main reaction device are controlled. Specifically, the following steps are included:

[0074] Obtain the production demand of calcium hypochlorite, preset the future reaction time series according to the production demand, and predict the particle agglomeration of the main reaction components of calcium hypochlorite at the future reaction time series through the particle agglomeration time series state prediction model, and output the particle agglomeration state segment of the main reaction components at the future reaction time series, marked as the future particle agglomeration state slice;

[0075] Obtain the specification variation index of different preset particle agglomeration particle sizes when the main reaction components produce calcium hypochlorite through big data, and assign the corresponding hot spot chromaticity components to each preset particle agglomeration particle size based on the specification variation index;

[0076] Introduce the Manhattan distance method to calculate the spatial correlation of the actual particle agglomeration particle size in each of the future particle agglomeration state slices, generate the covariance matrix of the future particle agglomeration state slices, and calculate the particle agglomeration particle size pattern in the covariance matrix based on the hot spot chromaticity components to obtain the Kriging equations of the particle agglomeration particle size pattern;

[0077] Simultaneously solve the Kriging equations to obtain the component interpolation weights of different actual particle agglomeration particle sizes in each future particle agglomeration state slice, and interpolate the hot spot chromaticity components corresponding to the hot spots to the actual particle agglomeration particle sizes in each particle agglomeration particle size pattern according to the component interpolation weights to generate the state hot spot chromaticity map of the actual particle agglomeration;

[0078] Obtain the normal particle agglomeration state of calcium hypochlorite through the production demand, preset the normal hot spot color gamut interval according to the normal particle agglomeration state. If the current state hot spot chromaticity shown in the state hot spot chromaticity map of the actual particle agglomeration cannot be found within the normal hot spot chromaticity interval, control the stirring rate of the calcium hypochlorite main reaction device and open the air release valve.

[0079] It should be noted that when the reaction between lime milk and chlorine gas is incomplete, it is easy to cause local or global particle agglomeration in the reactants of bleaching powder concentrate. The reason for the particle agglomeration may be insufficient stirring or over-stirring. Improper stirring makes the spatial position distribution of particle agglomeration uneven. Therefore, it is necessary to further make reasonable stirring control of the main reaction device of bleaching powder concentrate by monitoring the particle agglomeration condition of the main reaction components. For this purpose, this method can further predict the reaction state of the future reaction time series by using the particle agglomeration time series state prediction model, so as to know the particle agglomeration situation of the main reaction of bleaching powder concentrate in the future reaction time series, that is, the future particle agglomeration state slice. This future particle agglomeration state slice shows the possible particle agglomeration layout on the surface and inside of the main reaction components in the device. Since there are size differences in the particle agglomeration particle size specifications generated during the reaction process, for example, larger agglomerated particles appear in a certain local area, it means that the reaction degree between lime milk and chlorine gas in this local area is relatively serious, resulting in an increase in the particle size of the agglomerated particles. Therefore, for the abnormal reaction display in this area, a hot spot chromaticity component with a higher variation index should be used for display. Therefore, this method assigns the hot spot chromaticity components of the corresponding particle size area by obtaining the specification variation index of different preset particle agglomeration particle sizes when the main reaction components produce bleaching powder concentrate. The specification variation index represents the severity of different particle sizes generated by the main reaction components, so as to make the distribution density and specification display of the main reaction particle agglomeration in the device more accurate, and further help to improve the accuracy and rationality of the stirring control of the main reaction device of bleaching powder concentrate for different local areas.

[0080] It should be noted that the particle sizes of the agglomerated particles usually show an irregular spatial distribution among the main reaction components of calcium hypochlorite. However, the existing main reaction devices for calcium hypochlorite do not have technical methods to display the hot spot distribution in space, making it difficult for operators to quickly know the areas where agglomerated particles are likely to be generated inside and the spatial locations of particles of different specifications as the main reaction components proceed, thus unable to accurately control the stirring of the device to eliminate the generation of these agglomerated particles. In response to this, this method calculates the spatial correlation of the actual particle agglomeration particle size, generates the covariance matrix of the future particle agglomeration state slices, and this covariance matrix clearly describes the interdependent change trends of the agglomeration particle sizes of particles with similar sizes in the main reactant space, thereby enabling the subsequent display of hot spot chromaticity to be more coherent, and then effectively improving the global induction of the hot spot distribution in the space of the same particle size. Then, the preset hot spot chromaticity components of the specification variation index calculate the particle agglomeration particle size pattern in the covariance matrix to obtain the Kriging equations of the particle agglomeration particle size pattern; by simultaneously solving the Kriging equations, the interpolation weight components of different actual particle agglomeration particle sizes in each future particle agglomeration state slice can be obtained, so as to determine the contribution of each hot spot chromaticity component to the various actual particle agglomeration particle sizes in the particle agglomeration particle size pattern. Therefore, the corresponding interpolation of the hot spot chromaticity components can be carried out according to this weight, and finally a state hot spot chromaticity map showing the actual particle agglomeration spatial distribution is generated, revealing the distribution density and size of the agglomerated particles. If the current state hot spot chromaticity shown in the state hot spot chromaticity map of the actual particle agglomeration cannot be found within the normal hot spot chromaticity range, it indicates that the current particle agglomeration distribution does not conform to the normal situation of calcium hypochlorite preparation and may be affected by improper stirring rate. Therefore, control the stirring rate of the main reaction device for calcium hypochlorite. At the same time, when incomplete reaction causes particle agglomeration, unreacted chlorine or oxygen may accumulate, resulting in pressure increase. Therefore, synchronously control the opening of the relief valve to discharge excess chlorine or oxygen. Through this method, hot spot visualization analysis of the spatial distribution can be carried out based on the particle agglomeration phenomenon in the calcium hypochlorite reaction preparation process, so as to reasonably control the stirring rate and the timing of gas release of the main reaction device, effectively improve the accuracy of the calcium hypochlorite reaction preparation, eliminate the occurrence of particle agglomeration phenomenon to the greatest extent, and at the same time avoid the escape of excess chlorine, reduce the overpressure risk coefficient in the preparation process, and make the calcium hypochlorite preparation safer and more efficient.

[0081] More specifically, for constructing a gas generation quantization hash table when side reactions occur, using the gas generation quantization hash table to perform hash accumulation on the gases generated when side reactions occur in the main reaction components, and analyzing and controlling the reaction temperature of the device according to the hash accumulation result, specifically including the following steps:

[0082] Construct an initial gas quantification hash table, perform hash balancing quantification calculation on the reaction variable monitoring data based on the side chemical reaction balancing formula to obtain gas generation molar hash key-value pairs, analyze and store the gas generation molar hash key-value pairs in the initial gas quantification hash table to obtain a gas generation quantification hash table when side reactions occur;

[0083] When the air release valve of the bleaching powder main reaction device is in the open state, split the reaction variable monitoring data into N data blocks based on consecutive established monitoring nodes of the bleaching powder main reaction device, and construct a cumulative leaf node for each data block;

[0084] Perform hash queries on each cumulative leaf node through the gas generation quantification hash table to obtain the gas generation quantification hash values of each cumulative leaf node;

[0085] Starting from a certain cumulative leaf node, combine the adjacent two gas generation quantification hash values to generate a new hash combination for this cumulative leaf node, and use the gas generation quantification hash table to calculate the hash value of the new hash combination again to form a new gas generation quantification hash value;

[0086] Use the new gas generation quantification hash value as the cumulative leaf node of the upper layer, and continuously repeat the steps of combining adjacent gas generation quantification hash values and recalculating the hash value until a unique root hash value is generated, obtaining the gas cumulative root hash value when side reactions occur;

[0087] Identify the gas cumulative root hash value through the chemical reaction knowledge graph to obtain the temperature gradient that causes the generation of the gas cumulative root hash value, and construct it as the first temperature gradient curve;

[0088] Obtain the ideal temperature gradient for bleaching powder preparation, construct it as the second temperature gradient curve, calculate the slope difference between the first temperature gradient curve and the second temperature gradient curve to obtain the slope difference value, and control the internal reaction temperature of the bleaching powder main reaction device according to the slope difference value.

[0089] It should be noted that under humid or high-temperature conditions, the calcium hypochlorite generated by the reaction of lime milk and chlorine gas will undergo side reactions and decompose, generating a large amount of oxygen and chlorine gas. This can easily cause the pressure inside the main reaction device of bleaching powder concentrate to rise sharply, leading to device explosion or high-pressure damage, increasing the maintenance cost of the device and the operation risk coefficient. Therefore, it is extremely important to accurately and reasonably control the internal reaction temperature of the device. In response to this, this method quantifies the large amount of oxygen and chlorine gas generated by the decomposition of side reactions through the system, thereby constructing a gas generation quantification hash table. By querying this hash table, the specific amounts of oxygen and chlorine gas generated under different reaction variable monitoring data can be obtained, making the subsequent hash accumulation more accurate and able to truly restore the quantification effect of the gas gradually generating and increasing pressure inside the main reaction device of bleaching powder concentrate. Then, the reaction variable monitoring data is split into N cumulative leaf nodes to simplify the hash accumulation scale, making the cumulative decomposition of side reaction gases based on reaction variable monitoring data faster and more efficient, reducing unnecessary system hash operation steps, and improving verification flexibility. Next, the hash values of gas generation quantification corresponding to the leaf nodes of each reaction variable monitoring data are queried using this hash table, and the adjacent two values are combined and the cumulative leaf nodes of the upper layer are continuously updated, so as to achieve the gradual aggregation effect of oxygen or chlorine gas decomposed by side reactions inside the device as the reaction variable monitoring data is output. The finally presented system tree structure is the accumulation amount of oxygen or chlorine gas completely decomposed by side reactions inside the device, that is, the root hash value of gas accumulation when side reactions occur. Based on this gas accumulation root hash value, the unreasonable temperature cause leading to its generation can be known, and then the unreasonable temperature can be improved according to the ideal temperature. Through this method, the specific amounts of oxygen or chlorine gas decomposed by side reactions under reaction variable monitoring conditions can be quantified by hashing, and systematic hash accumulation is performed on these specific amounts to determine and reasonably control the temperature, improve the high-temperature conditions of the internal reaction of the device, reduce the phenomenon of excessive closed pressure of the device caused by the escape of chlorine or oxygen gas, improve the safety of the device reaction to prepare bleaching powder concentrate, and at the same time replace the inaccurate steps of the traditional manual judgment of internal pressure, making the temperature control of the main reaction device of bleaching powder concentrate more intelligent and improving the temperature control accuracy.

[0090] More specifically, for constructing the initial gas quantification hash table, hash balancing quantification calculation is performed on the reaction variable monitoring data based on the side chemical reaction balancing formula to obtain gas generation molar hash key-value pairs. The gas generation molar hash key-value pairs are analyzed and stored in the initial gas quantification hash table to obtain the gas generation quantification hash table when side reactions occur, which specifically includes the following steps:

[0091] Obtain the chemical reaction knowledge graph based on the big data network, identify the main reaction components through the chemical reaction knowledge graph, and output the reaction characteristics when the main reaction components are incompletely dissolved;

[0092] Introduce the gas side reaction molar law, use the reaction characteristics as the indexing rule, build the storage key table structure of the gas side reaction molar law based on the indexing rule, and generate the initial gas quantization hash table;

[0093] Obtain the balanced formula of the side chemical reaction when the main reaction component is incompletely dissolved. Introduce the cryptographic hash algorithm, and use the cryptographic hash algorithm to perform balanced quantization calculation on the reaction variable monitoring data in the side reaction chemical formula to obtain the gas generation molar hash key value pairs when the main reaction component is incompletely dissolved under different reaction variable monitoring data conditions;

[0094] If the index position of the gas quantization initial hash table for the gas generation molar hash key value pair is empty, directly store the gas generation molar hash key value to the current index position to obtain a type of gas hash quantization;

[0095] If the gas quantization initial hash table maps the gas generation molar hash key value pair to the same index, perform quadratic probing to find the new storage position of the gas generation molar hash key value to obtain a second type of gas hash quantization;

[0096] Perform dynamic update adjustment on the gas quantization initial hash table based on the first type of gas hash quantization and the second type of gas hash quantization to obtain the gas generation quantization hash table when side reactions occur.

[0097] It should be noted that for the convenience of quickly querying gas quantization, this method uses the method of constructing a hash table to achieve the hash quantization indexing effect of the gas volume under different reaction variable monitoring data conditions, so that the gas volume accumulated by subsequent hashing is more clearly and efficiently reflected, and thus can adapt to the gas calculation response rate of real-time monitoring of reaction variables. When side reactions of incomplete dissolution occur between lime milk and chlorine gas under certain concentration and acidity conditions, the amount of oxygen or chlorine gas generated is constant, which follows the reaction characteristics and the gas side reaction molar law. Therefore, for the quantization of the gas generation amount, this method identifies and obtains the reaction characteristics when the main reaction component is incompletely dissolved, which is the characteristic premise for the side reaction decomposition of oxygen or chlorine gas in the preparation of bleaching powder essence, and can ensure the indexing accuracy of the conversion of the gas generation amount. Then, use the indexing rule of the reaction characteristics to construct the initial gas quantization hash table of the gas side reaction molar law, so that the quantization of the gas can follow the gas side reaction molar law for quantization indexing based on the reaction characteristics, avoiding quantization conflicts and errors, and maintaining the chemical specificity of gas generation to the greatest extent. Then, further combine the hash algorithm and the balanced formula of the side chemical reaction when the main reaction component is incompletely dissolved to perform balanced quantization on the reaction variable monitoring data, generate gas generation molar hash key value pairs, and store these key value pairs in the constructed initial gas quantization hash table.

[0098] It should be noted that if the index position of the gas quantification initial hash table for the gas generation molar hash key-value pair is empty, it indicates that under the reaction variable monitoring data, this gas can find the corresponding hash quantity value through the stoichiometric quantification that follows the gas side reaction molar law. Therefore, directly store the gas generation molar hash key to the current index position; if it is mapped to the same index, it means that the corresponding accurate hash quantity value cannot be found, and there are quantification conflicts or errors, resulting in inaccurate gas quantification. Therefore, it is necessary to continue to search for a suitable hash quantity value to correspond to the stoichiometric quantity of this gas that follows the gas side reaction molar law. Through this method, a hash table framework index can be constructed using the gas side reaction molar law and the reaction characteristics when the main reaction components are incompletely dissolved to store the hash key-value pairs of the reaction variable monitoring data after stoichiometric quantification, thereby constructing a unit quantification query table for subsequent internal gas hash accumulation in the device, which can ensure the quantification accuracy of gas aggregation per unit time, improve the estimation accuracy of gas over-decomposition and escape caused by high-temperature conditions, and effectively guarantee the temperature control performance of the calcium hypochlorite main reaction device.

[0099] The second aspect of the present invention provides a control system for the calcium hypochlorite main reaction device based on reaction variable monitoring, as Figure 3 shown. The control system for the calcium hypochlorite main reaction device includes a memory 31 and a processor 32. A control method program for the calcium hypochlorite main reaction device based on reaction variable monitoring is stored in the memory 31. When the control method program for the calcium hypochlorite main reaction device is executed by the processor 32, the steps of any of the control methods for the calcium hypochlorite main reaction device are implemented.

[0100] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A control method for a bleaching powder main reaction device based on reaction variable monitoring, characterized in that: The following steps are involved: Acquire the reaction variable monitoring data of bleaching powder, and perform state random deduction on the reaction variable monitoring data by constructing the standard reaction random evolution equation of bleaching powder to obtain the simulated reaction state of bleaching powder in the device; Calculate the state transition value of particle agglomeration in the bleaching powder essence reaction according to the simulated reaction state adjacency topology on different predetermined monitoring nodes, so as to construct a particle agglomeration time series state prediction model of the bleaching powder essence reaction; The future state of particle agglomeration is predicted by the particle agglomeration time series state prediction model, and the state hot spot chromaticity diagram of actual particle agglomeration is interpolated according to the hot spot chromaticity of the prediction result. The stirring rate and the air release valve of the main reaction device of bleaching powder are controlled based on the state hot spot chromaticity diagram. A gas generation quantification hash table is constructed when a side reaction occurs. The gas generation quantification hash table is used to perform hash accumulation on the gas generated when a side reaction occurs in the main reaction component. The reaction temperature of the control device is analyzed based on the hash accumulation result.

2. A bleaching powder main reaction device control method based on reaction variable monitoring according to claim 1, characterized in that: The step of obtaining the reaction variable monitoring data of the bleaching powder, performing state random deduction on the reaction variable monitoring data by constructing a standard reaction random evolution equation of the bleaching powder, and obtaining a simulated reaction state of the bleaching powder in the device specifically includes the following steps: Obtaining a standard reaction process for bleaching powder, extracting the main reaction components of bleaching powder through the standard reaction process, and obtaining a reference reaction concentration and an accurate rate constant of the main reaction components when producing qualified bleaching powder; Based on the standard reaction process, the reaction kinetics formula of the main reaction component is searched in the big data network, and the benchmark reaction concentration and the accurate rate constant are dynamically calculated by the reaction kinetics formula to obtain the standard reaction tendency function; Obtain the reaction variable monitoring data and reaction process of the main reaction device of bleaching powder, introduce the molecular random evolution model, perform evolution description of the reaction process in the molecular random evolution model based on the standard reaction process, and output the standard reaction random evolution equation of bleaching powder; Solving the standard reaction random evolution equation through the standard reaction tendency function to obtain the Poisson trigger time step of the next main reaction, and accumulating and calculating the reaction rate distribution of the main reaction components based on the reaction variable monitoring data. During the accumulation process, the timing of the standard reaction random evolution equation is advanced according to the Poisson trigger time step of the next main reaction. After the reaction termination step is reached, a series of evolution reactions of the main reaction components under the cumulative rate distribution are output, and the state of the standard reaction random evolution equation is updated according to the evolution reaction to obtain the simulated reaction state of the bleaching powder in the device.

3. A bleaching powder main reaction device control method based on reaction variable monitoring according to claim 1, characterized in that: The calculation of the state transition value of particle agglomeration in the bleaching powder reaction according to the simulated reaction state adjacency topology on different predetermined monitoring nodes to construct a particle agglomeration time series state prediction model for the bleaching powder reaction specifically includes the following steps: Obtaining a preset monitoring strategy for monitoring reaction variables of the bleaching powder main reaction device, and extracting continuous predetermined monitoring nodes of the bleaching powder main reaction device through the preset monitoring strategy; The simulated reaction state of the main reaction device of the bleaching powder essence at each predetermined monitoring node is obtained, which is defined as the current simulated reaction state, and the state boundary weight of the bleaching powder essence transferred from the current simulated reaction state at the current predetermined monitoring node to the next predetermined monitoring node is calculated according to the reaction variable monitoring data distribution; Taking the current simulated reaction state as the state adjacent loop node, connecting each state adjacent loop node based on the state boundary weight to construct a state transition adjacency graph of the current simulated reaction state, and extracting the state topology layout of the bleaching powder essence reaction through the state transition adjacency graph; Based on the big data network, the impending reaction conditions of the agglomeration phenomenon of bleaching powder particles are obtained. According to the impending reaction conditions, a reaction state evaluation system for the agglomeration of particles in the bleaching powder reaction is constructed. The Bellman equation is introduced, and the value of each transfer state in the update state topology layout is evaluated in the Bellman equation based on the reaction state evaluation system to obtain the value function of the transfer state. The maximum critical standard of the reaction state evaluation system is extracted, and the value function threshold is preset based on the maximum critical standard. If the value function is greater than the value function threshold, the evaluation and update iteration steps of the value function of the above transfer state are repeated until it is less than the value function threshold. The value function matrix is ​​obtained, and a particle agglomeration time series state prediction model of the bleaching powder reaction is constructed based on the value function matrix training.

4. A bleaching powder main reaction device control method based on reaction variable monitoring according to claim 1, characterized in that: The method predicts the future particle agglomeration state through the particle agglomeration time series state prediction model, interpolates the state hot spot chromaticity diagram of the actual particle agglomeration according to the prediction result spatial chromaticity, and controls the stirring rate and the air release valve of the bleaching powder main reaction device based on the state hot spot chromaticity diagram, specifically including the following steps: Obtaining the production demand of bleaching powder, presetting the future reaction sequence according to the production demand, predicting the agglomeration of the main reaction component particles of the bleaching powder in the future reaction sequence through the particle agglomeration sequence state prediction model, and outputting the particle agglomeration state fragment of the main reaction component in the future reaction sequence, which is marked as the future particle agglomeration state slice; The specification variation index of different preset particle agglomerate sizes when the main reaction components are used to produce bleaching powder is obtained through big data, and the hot spot chromaticity component corresponding to each preset particle agglomerate size is assigned based on the specification variation index; The Manhattan distance method is introduced to calculate the spatial correlation of the actual particle agglomeration size in each of the future particle agglomeration state slices, and the covariance matrix of the future particle agglomeration state slices is generated. The particle agglomeration size pattern is calculated in the covariance matrix based on the hot spot chromaticity component, and the Kriging equation group of the particle agglomeration size pattern is obtained; The Kriging equations are solved simultaneously to obtain component interpolation weights of different actual particle agglomeration particle sizes in each future particle agglomeration state slice, and the hot spot chromaticity component corresponding to the hot spot is interpolated to the actual particle agglomeration particle size in each particle agglomeration particle size pattern according to the component interpolation weight to generate a state hot spot chromaticity diagram of the actual particle agglomeration; The normal particle agglomeration state of bleaching powder is obtained through production needs, and a normal hot spot color range is preset according to the normal particle agglomeration state. If the current state hot spot chromaticity displayed by the hot spot chromaticity diagram of the actual particle agglomeration state cannot be queried within the normal hot spot chromaticity range, the stirring rate of the bleaching powder main reaction device is controlled and the air release valve is opened.

5. The control method of a bleaching powder main reaction device based on reaction variable monitoring according to claim 1 is characterized in that: The method of constructing a gas generation quantification hash table when a side reaction occurs, using the gas generation quantification hash table to perform hash accumulation on the gas generated when a side reaction occurs in the main reaction component, and analyzing the reaction temperature of the control device according to the hash accumulation result specifically includes the following steps: Constructing a gas quantization initial hash table, performing hash balancing quantization calculation on the reaction variable monitoring data based on the side chemical reaction balancing formula, obtaining the gas generation mole hash key-value pairs, analyzing and storing the gas generation mole hash key-value pairs in the gas quantization initial hash table, and obtaining the gas generation quantization hash table when the side reaction occurs; When the air release valve of the bleaching powder main reaction device is in an open state, the reaction variable monitoring data is split into N data blocks based on the continuous established monitoring nodes of the bleaching powder main reaction device, and a cumulative leaf node is constructed according to each data block; Perform a hash query on each cumulative leaf node through the gas generation quantization hash table to obtain the gas generation quantization hash value of each cumulative leaf node; Starting from a certain cumulative leaf node, two adjacent gas-generated quantized hash values ​​are combined to generate a new hash combination of the cumulative leaf node, and the hash value of the new hash combination is calculated again using the gas-generated quantized hash table to form a new gas-generated quantized hash value; The newly generated quantized hash value of the gas is used as the cumulative leaf node of the previous layer, and the steps of combining the quantized hash values ​​of the adjacent gases and recalculating the hash values ​​are continuously repeated until a unique root hash value is generated, and the cumulative root hash value of the gas when the side reaction occurs is obtained; Identify the gas cumulative root hash value through a chemical reaction knowledge graph to obtain a temperature gradient that causes the gas cumulative root hash value to be generated, and construct a first temperature gradient curve; The current temperature gradient of the bleaching powder main reaction device is obtained, a second temperature gradient curve is constructed, the slope difference between the first temperature gradient curve and the second temperature gradient curve is calculated to obtain the slope difference, and the internal reaction temperature of the bleaching powder main reaction device is controlled according to the slope difference.

6. A control method for a bleaching powder main reaction device based on reaction variable monitoring according to claim 5, characterized in that: The gas quantization initial hash table is constructed, and the reaction variable monitoring data is subjected to hash balancing quantization calculation based on the side chemical reaction balancing formula to obtain the gas generation mole hash key-value pairs, and the gas generation mole hash key-value pairs are analyzed and stored in the gas quantization initial hash table to obtain the gas generation quantization hash table when the side reaction occurs, specifically including the following steps: Obtain chemical reaction knowledge graph based on big data network, identify main reaction components through chemical reaction knowledge graph, and output reaction characteristics when main reaction components are incompletely dissolved; The Moore's law of gas side reactions is introduced, and the reaction characteristics are used as index rules. Based on the index rules, the storage key table architecture of the Moore's law of gas side reactions is constructed to generate the gas quantization initial hash table. Obtain the side chemical reaction balancing formula when the main reaction component is incompletely dissolved, introduce the encrypted hash algorithm, use the encrypted hash algorithm to perform balancing quantitative calculations on the reaction variable monitoring data in the side reaction chemical formula, and obtain the gas generation mole hash key-value pairs when the main reaction component is incompletely dissolved under different reaction variable monitoring data conditions; If the index position of the gas-generated mole hash key-value pair in the gas quantization initial hash table is empty, the gas-generated mole hash key value is directly stored at the current index position to obtain a type of gas hash quantization; If the gas quantization initial hash table maps the gas generated mole hash key-value pair to the same index, the secondary detection searches for the new storage location of the gas generated mole hash key value to obtain the second type of gas hash quantization; Based on the first-class gas hash quantization and the second-class gas hash quantization, the gas quantization initial hash table is dynamically updated and adjusted to obtain the gas generation quantization hash table when the side reaction occurs.

7. A control system for a bleaching powder main reaction device based on reaction variable monitoring, characterized in that: The bleaching powder essence main reaction device control system includes a memory and a processor, wherein the memory stores a bleaching powder essence main reaction device control method program based on reaction variable monitoring, and when the bleaching powder essence main reaction device control method program is executed by the processor, the bleaching powder essence main reaction device control method steps as described in any one of claims 1 to 6 are implemented.

Images