CN-121971967-A - Intelligent spray control processing method and equipment based on temperature field space-time characteristics

CN121971967ACN 121971967 ACN121971967 ACN 121971967ACN-121971967-A

Abstract

The application relates to the technical field of intelligent control, and provides an intelligent spray control processing method and equipment based on temperature field space-time characteristics, which are used for improving response timeliness and regulation accuracy of a spray control system. The method comprises the steps of performing dimension mapping processing on original temperature field data of a target area to generate temperature field related data, merging time variation trend features and spatial distribution features of the temperature field related data to generate temperature field space-time features, importing a gas concentration prediction model to output a gas concentration prediction result, determining average temperature gradient strength through context connectivity of the temperature field data between adjacent spatial coordinate marks, determining a spatial deviation distance through deviation situation comparison of preset standard temperature field data and the temperature field related data, generating spray control strategy data through merging the temperature field space-time features, the average temperature gradient strength, the spatial deviation distance and other multiple parameters, transmitting the spray control strategy data to a spray control system, and performing double situation collaborative correction.

Inventors

DENG JUN
GUO HAITAO
SUN QIUSHENG
ZHANG XIAOXIAO
LIU YANG
LIU ZHIHENG
CHEN LEI
HOU YUE
CHEN DONGXU

Assignees

北京朝阳环境集团有限公司

Dates

Publication Date: 20260505
Application Date: 20251227

Claims (10)

1. An intelligent spray control processing method based on temperature field time-space characteristics is characterized by comprising the following steps: Performing dimension mapping processing on the original temperature field data of the target area according to the acquisition time sequence and the acquisition position to generate temperature field related data with a time sequence mark and a space coordinate mark; Extracting time variation trend characteristics of a temperature field based on the continuous relevance of the time sequence marks, and extracting spatial distribution characteristics of the temperature field based on the regional relevance of the spatial coordinate marks; Generating temperature field space-time characteristics according to the time variation trend characteristics and the spatial distribution characteristics, importing a preset gas concentration prediction model, and outputting a gas concentration prediction result of the target area through characteristic analysis and trend deduction; Traversing temperature field data corresponding to different space coordinate marks in the temperature field associated data, identifying a space change situation of the temperature field through context connectivity of the temperature field data between adjacent space coordinate marks, and determining average temperature gradient strength based on global coverage and global uniformity of the space change situation; Comparing the deviation situation of the preset standard temperature field data aligned according to the time sequence mark and the space coordinate mark with the temperature field related data to generate deviation state monitoring information, and determining a space deviation distance according to the space spreading rule of the deviation state monitoring information; And transmitting a control logic instruction corresponding to the control strategy data to a control system, and carrying out double-situation collaborative correction of the control logic instruction based on the updated situation of the temperature field associated data and the evolution situation of the gas concentration prediction result.
2. The method of claim 1, wherein the generating the spatial distribution feature according to the time variation trend feature to generate a temperature field space-time feature and importing the temperature field space-time feature into a preset gas concentration prediction model, and outputting a gas concentration prediction result of the target area through feature analysis and trend deduction comprises: Generating a feature alignment matrix based on the time sequence mark continuity of the time change trend feature and the space coordinate mark relevance of the space distribution feature, determining a dynamic adaptation incidence mapping relation between the time sequence dimension of the time change trend feature and the space dimension of the space distribution feature through the feature alignment matrix, and generating a temperature field space-time feature according to the dynamic adaptation incidence mapping relation; Calculating the feature contribution degree of each feature dimension in the temperature field space-time feature based on an information gain rate algorithm, determining a key feature dimension by combining the feature contribution degree, and eliminating the feature components with redundancy association in the key feature dimension through mutual information entropy calculation to obtain a temperature field space-time optimization feature; introducing the temperature field space-time optimization features into a preset gas concentration prediction model to execute multi-scale hierarchical analysis, and disassembling to form hierarchical feature components corresponding to different influence hierarchies of the temperature field space-time variation; based on the time sequence marks and the space coordinate marks carried by the hierarchical feature components, performing multi-stage trend deduction by fusing the time sequence evolution rule and the space propagation rule of the time-space change of the temperature field, and generating gas concentration trend prejudging results covering different time windows and space areas; Carrying out reliability quantification on the gas concentration trend pre-judging results of each stage by introducing a confidence evaluation strategy, carrying out feature fusion and resolving feature conflict on the gas concentration trend pre-judging results of different stages based on the quantified results, and generating a unified gas concentration predicting result of a target area with confidence parameters; And establishing a bidirectional association index between the gas concentration prediction result and the key feature dimension, enabling the gas concentration prediction result to reversely trace back to a corresponding temperature field space-time feature component through the bidirectional association index, and enabling the temperature field space-time feature component to inquire a corresponding gas concentration prediction result influence weight through the bidirectional association index.
3. The method of claim 1, wherein traversing temperature field data corresponding to different spatial coordinate markers in the temperature field-related data, identifying spatially-varying situations of the temperature field by context-connectivity of temperature field data between adjacent spatial coordinate markers, determining an average temperature gradient strength based on global coverage and global uniformity of the spatially-varying situations, comprises: traversing corresponding temperature field data based on the region topological relation of the space coordinate marks in the temperature field associated data, and extracting dynamic state characteristics of the temperature field data corresponding to each space coordinate mark under the full-time dimension; Calculating time sequence similarity between dynamic state features corresponding to adjacent space coordinate marks, correcting the time sequence similarity by combining with physical distance weights between the adjacent space coordinate marks, judging context linking of temperature field data between the adjacent space coordinate marks based on a corrected time sequence similarity threshold, judging linking continuity when the corrected time sequence similarity is higher than the similarity threshold, and judging linking breakage when the corrected time sequence similarity is lower than the similarity threshold; generating a temperature field space association network by taking a space coordinate mark as a network node and taking the corrected time sequence similarity as network edge weight, and identifying a space change situation of a temperature field by calculating node degree centrality, edge weight distribution variance and network clustering coefficient of the temperature field space association network; dividing the global range into a plurality of heterogeneous subareas based on the zoning result of the space change situation, and fusing temperature field data and time sequence marking information corresponding to all space coordinate marks in each heterogeneous subarea to calculate a space temperature gradient vector, wherein the space temperature gradient vector comprises a gradient direction and a gradient amplitude; determining region weights according to the area ratio of the regions with different protons and the information quantity ratio of the temperature data, and carrying out weighted aggregation on the space temperature gradient vectors of the regions with different protons based on the region weights to obtain initial average temperature gradients; Adopting a self-adaptive grid encryption algorithm to adjust the division granularity of heterogeneous subareas, repeatedly executing the steps of space temperature gradient vector calculation and weighted aggregation until the initial average temperature gradient difference value obtained by two adjacent times is smaller than a preset convergence threshold value, and determining the initial average temperature gradient after convergence as the average temperature gradient strength; And constructing a bidirectional association mapping table of the average temperature gradient strength and the space change situation, determining the value ranges of the average temperature gradient strength under different types of space change situations through the bidirectional association mapping table, and marking the heterogeneous subarea division granularity and area weight distribution rules corresponding to each value range so as to realize the mutual traceability of the average temperature gradient strength and the space change situation.
4. The method of claim 1, wherein comparing the deviation situation of the preset standard temperature field data aligned according to the time sequence mark and the space coordinate mark with the temperature field related data to generate deviation state monitoring information, and determining a space deviation distance according to a space spreading rule of the deviation state monitoring information comprises: Aligning the preset standard temperature field data according to the time sequence mark and the space coordinate mark, so that the corresponding relation between the preset standard temperature field data and the temperature field associated data is kept consistent in both time sequence dimension and space dimension; Comparing the aligned preset standard temperature field data with the temperature field related data in a time sequence and space coordinate mark by space coordinate mark, extracting difference characteristics between the aligned preset standard temperature field data and the temperature field related data, and generating deviation state monitoring information based on the difference characteristics, wherein the deviation state monitoring information comprises deviation direction and deviation degree data corresponding to each time sequence and each space coordinate mark; Carrying out time sequence tracing on the deviation state monitoring information, analyzing the spatial position change of a deviation area under different time sequences, and identifying the spatial spreading rule of the deviation state, wherein the spatial spreading rule comprises a spreading direction, a spreading speed and a spreading range expanding trend; Constructing a deviation spreading path model comprising a deviation starting position, a spreading passing position and a potential spreading ending position based on the space spreading rule, calculating the space distance from the deviation starting position to the boundary of the current deviation area, correcting the space distance by combining the spreading speed and the spreading range expansion trend, and determining the corrected space distance as a space deviation distance; and correlating the spatial deviation distance, the deviation direction and the deviation degree data to generate a deviation characteristic comprehensive data table.
5. The method of claim 4, wherein the performing timing trace on the deviation status monitoring information, analyzing the spatial position change of the deviation area under different timings, and identifying the spatial spreading rule of the deviation status, includes: Generating a deviation state monitoring time sequence of full-time dimension based on the time sequence mark carried by the deviation state monitoring information, so that each data node in the deviation state monitoring time sequence and the corresponding acquisition time sequence form one-to-one correspondence; Extracting space coordinate marks and deviation degree data contained in deviation state monitoring information corresponding to each time sequence node in the deviation state monitoring time sequence, screening out a space coordinate mark set meeting deviation judgment conditions according to a preset deviation degree threshold value, defining a space area corresponding to the space coordinate mark set as a deviation area under each time sequence, and determining a space contour boundary of each deviation area; Sequentially extracting a space contour boundary and a space coordinate mark set of a deviation area corresponding to two adjacent time sequence nodes based on the time sequence continuity of the deviation state monitoring time sequence, calculating a space overlapping area occupation ratio and a center coordinate offset of the deviation area under the adjacent time sequence, and quantifying the space position change amplitude of the deviation area between the adjacent time sequences by combining the space overlapping area occupation ratio and the center coordinate offset; Generating a space evolution track model of the deviation area by taking a time sequence as a horizontal axis and space position parameters of the deviation area as a vertical axis, fitting the space evolution track model of the deviation area to obtain a space movement track curve of the deviation area, and determining the spreading direction of the deviation state based on tangential direction distribution characteristics of the space movement track curve; Calculating the ratio of the spatial position change amplitude of the adjacent time sequence deviation area to the time interval to obtain the instantaneous spreading speed, carrying out moving average treatment on the instantaneous spreading speed in the whole time sequence range to obtain the average spreading speed, and determining the spreading speed characteristic by combining the fluctuation amplitude of the instantaneous spreading speed; And counting the number of the space coordinate marks corresponding to the deviation areas and the area of the covered areas under each time sequence, analyzing the change trend of the covered areas along with the time sequence to obtain the spreading range expansion trend, and fusing the spreading direction, the spreading speed characteristic and the spreading range expansion trend to obtain the space spreading rule of the deviation state.
6. The method of claim 1, wherein the performing feature-dependent deduction with the temperature field spatiotemporal feature as a dependent target, fusing the gas concentration prediction result, the average temperature gradient strength, and the spatial deviation distance, generates the injection strategy data, comprising: Performing dimension alignment processing on the gas concentration prediction result, the average temperature gradient strength and the space deviation distance by taking a time sequence mark and a space coordinate mark carried by the space-time characteristic of the temperature field as references to generate a unified characteristic set containing multi-source characteristics, wherein each characteristic component in the unified characteristic set is associated with a corresponding time sequence mark and space coordinate mark; Generating a feature correlation matrix based on the unified feature set, determining linear correlation degrees of temperature field space-time features, the gas concentration prediction result, the average temperature gradient strength and the space deviation distance by calculating pearson correlation coefficients among feature components in the feature correlation matrix, supplementing nonlinear correlation degree assessment by combining a mutual information value, and fusing the linear correlation degrees with the nonlinear correlation degrees to generate feature correlation weights; Weighting and fusing all feature components in the unified feature set according to the feature association weights to generate a multi-dimensional fusion feature vector, importing the multi-dimensional fusion feature vector into a preset strategy deduction model, and setting constraint conditions by combining a temperature field regulation threshold value and a gas concentration safety threshold value of a target area; carrying out situation matching deduction on the multidimensional fusion feature vector based on the constraint condition, identifying the corresponding spraying control demand type under the space-time feature scene of different temperature fields, and generating an initial spraying control strategy set comprising spraying control intensity parameters, spraying control range parameters and spraying control time sequence parameters; And calculating an average temperature gradient alleviation expected value and a spatial deviation correction expected value corresponding to each strategy in the initial spray control strategy set, carrying out effect evaluation on each initial spray control strategy by combining the improvement expected value of the gas concentration prediction result, and screening out the initial spray control strategy with the maximum evaluation value as the spray control strategy data.
7. The method of claim 1, wherein transmitting the regulation logic command corresponding to the injection control strategy data to an injection control system, and performing a double-situation collaborative correction of the regulation logic command based on the updated situation of the temperature field related data and the evolution situation of the gas concentration prediction result, comprises: Converting the control logic instruction corresponding to the spray control strategy data according to the communication protocol of the spray control system, and transmitting the control logic instruction to the spray control system to finish the initial issuing of the control instruction; Collecting temperature field update data of a target area in real time, and extracting update situation characteristics of the temperature field associated data by combining the historical temperature field associated data, wherein the update situation characteristics comprise a temperature field change rate, a spatial diffusion range of the temperature field update area and time continuity; carrying out time sequence tracking on the gas concentration prediction result based on a preset time window, calculating deviation values of the gas concentration prediction result and actually collected gas concentration data under different time sequence nodes, and extracting evolution situation features of the gas concentration prediction result by combining the change trend of the deviation values, wherein the evolution situation features comprise deviation evolution directions, deviation accumulation amplitudes and deviation convergence trends; introducing updated situation characteristics of temperature field associated data and evolution situation characteristics of a gas concentration prediction result into a preset double-situation collaborative association model to calculate situation coupling degree, and determining regulation logic instruction correction priority under different coupling degree scenes; Based on the regulation logic instruction correction priority and the double-situation characteristic parameter, a corresponding correction rule base is created, wherein the correction rule comprises a regulation intensity correction coefficient, a regulation range correction threshold value and a determination standard of a regulation time sequence offset; And carrying out parameter adjustment on the currently issued regulation and control logic instruction according to the correction rule base, generating a corrected regulation and control logic instruction, and transmitting the corrected regulation and control logic instruction to the spray control system to update the regulation and control instruction.
8. The method of claim 7, wherein the step of importing the updated situation feature of the temperature field related data and the evolution situation feature of the gas concentration prediction result into a preset double-situation collaborative correlation model to calculate situation coupling degree, and determining the correction priority of the regulation logic instruction under different coupling degree scenes comprises the following steps: Extracting first characteristic attributes of temperature field change rate, spatial diffusion range and time continuity of a temperature field update region in update situation characteristics of temperature field association data, determining association influence dimensions of each first characteristic attribute and regulation logic instruction correction, extracting second characteristic attributes of deviation evolution direction, deviation accumulation amplitude and deviation convergence trend in evolution situation characteristics of a gas concentration prediction result, and establishing corresponding association relations between each second characteristic attribute and regulation logic instruction correction; Generating a characteristic association rule of a double-situation collaborative association model based on the spray control regulation and control demand information of the target area, wherein the characteristic association rule is used for defining a collaborative relationship and regulation and control influence weight distribution basis between each attribute of a temperature field updating situation characteristic and a gas concentration evolution situation characteristic; Introducing an updated situation characteristic of the temperature field associated data, an evolution situation characteristic of a gas concentration prediction result and a double situation characteristic correction influence associated map into a preset double situation collaborative association model, and mining collaborative association degree among the double situation characteristics through situation association analysis to generate situation coupling degree representing double situation collaborative association level; based on the corresponding relation between the double-situation collaborative association level and the correction effect in the historical spray control correction case, different situation coupling degree scenes are divided, and expected labels and response requirements of correction of the control logic instructions are adjusted under each situation coupling degree scene of the channel; and setting corresponding regulation and control logic instruction correction priority according to expected labels and response requirements of all situation coupling degree scenes.
9. A computer device, characterized in that it comprises a processor and a memory, wherein the memory stores a computer program, which when executed by the processor, causes the processor to perform the steps of the intelligent spray control processing method based on the temperature field spatiotemporal features of any of claims 1 to 8.
10. A computer readable storage medium, characterized in that the computer readable storage medium comprises a computer program for causing a computer device to execute the steps of the intelligent spray control processing method based on the temperature field spatiotemporal features of any of claims 1 to 8 when the computer program is run on the computer device.

Description

Intelligent spray control processing method and equipment based on temperature field space-time characteristics Technical Field The application belongs to the technical field of intelligent control, and particularly relates to an intelligent spray control processing method and equipment based on temperature field space-time characteristics. Background At present, in the flue gas denitration process of a waste incineration power plant, a control strategy based on fixed PID or simple feedforward-feedback is generally adopted to adjust the ammonia injection amount. However, due to complex garbage components and large heat value fluctuation, the combustion working condition of the boiler is unstable, and the flue gas parameters are severely changed. The existing control method seriously depends on real-time measurement value of the outlet NOx, has obvious hysteresis, and cannot respond to rapid change of combustion state in time, so that the control system is always in a 'passive response' state, and the following problems are caused: 1) NOx transient overproof risk, namely when load or fuel suddenly changes, NOx generation amount changes sharply, and the control system responds slowly, so that emission transient overproof is caused; 2) The excessive consumption of ammonia water, namely, in order to avoid the risk of exceeding the standard, operators often set higher ammonia spraying redundancy quantity, so that the consumption of the ammonia water is large, and the operation cost is high; 3) Equipment corrosion and blockage, namely ammonia escape caused by excessive ammonia injection, and ammonium bisulfate generated by reaction with sulfur trioxide, so that downstream equipment such as an air preheater and the like are blocked and corroded. Therefore, how to ensure the regulation precision and the response timeliness of the ammonia injection quantity control strategy is a technical problem which needs to be solved at present. Disclosure of Invention The application provides an intelligent spray control processing method and equipment based on temperature field space-time characteristics, which are used for improving the response timeliness and regulation accuracy of a spray control system. The embodiment of the application provides an intelligent spray control processing method based on temperature field time-space characteristics, which is applied to computer equipment and comprises the following steps: Performing dimension mapping processing on the original temperature field data of the target area according to the acquisition time sequence and the acquisition position to generate temperature field related data with a time sequence mark and a space coordinate mark; Extracting time variation trend characteristics of a temperature field based on the continuous relevance of the time sequence marks, and extracting spatial distribution characteristics of the temperature field based on the regional relevance of the spatial coordinate marks; Generating temperature field space-time characteristics according to the time variation trend characteristics and the spatial distribution characteristics, importing a preset gas concentration prediction model, and outputting a gas concentration prediction result of the target area through characteristic analysis and trend deduction; Traversing temperature field data corresponding to different space coordinate marks in the temperature field associated data, identifying a space change situation of the temperature field through context connectivity of the temperature field data between adjacent space coordinate marks, and determining average temperature gradient strength based on global coverage and global uniformity of the space change situation; Comparing the deviation situation of the preset standard temperature field data aligned according to the time sequence mark and the space coordinate mark with the temperature field related data to generate deviation state monitoring information, and determining a space deviation distance according to the space spreading rule of the deviation state monitoring information; And transmitting a control logic instruction corresponding to the control strategy data to a control system, and carrying out double-situation collaborative correction of the control logic instruction based on the updated situation of the temperature field associated data and the evolution situation of the gas concentration prediction result. An embodiment of the present application provides a computer device including a processor and a memory, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of the above method. Embodiments of the present application provide a computer readable storage medium comprising a computer program for causing a computer device to carry out the steps of the above-described method when said computer program is run on the computer device. The embodiment of the application