An automatic control method and system for solid waste incineration equipment

Abstract

The present application relates to solid waste incineration automatic control technical field, disclose a kind of solid waste incineration equipment automatic control method and system, the method includes: collecting solid waste composition real-time data in hearth, and initial multidimensional parameter set is obtained after pre-processing;The balance point of calorific value and moisture content is calculated by fuzzy reasoning;When the balance point exceeds threshold value, then predict parameter interaction, generate multidimensional control vector;Combining real-time incineration process index carries out data fusion and historical data matching, and obtains enhanced load distribution adjustment factor;Based on the weight comparison of adjustment factor and balance point, judge load balancing influence index, if the index exceeds compatible threshold value, then carry out matching correction;According to the revision result, update load distribution scheme, generate final operation strategy;Through feedback cycle mechanism, send control instruction and continuously optimize, realize closed-loop control.The present application can adaptively handle solid waste composition dynamic change, improve incineration efficiency and stability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic control of solid waste incineration, and particularly relates to an automatic control method and system for solid waste incineration equipment. BACKGROUND

[0002] At present, solid waste incineration treatment is an important part of modern urban environmental management, and is widely used in the treatment of household garbage and industrial waste, which is directly related to the comprehensive benefits of resource recycling, pollution control and energy utilization. This technology converts waste into usable energy through high-temperature incineration, which not only effectively reduces environmental pollution, but also has significant social and economic value. However, due to the diverse sources of solid waste, its composition is extremely complex and variable, involving key parameters such as calorific value, moisture content, chemical composition, etc. The dramatic fluctuations of these parameters bring great technical challenges to the stable, efficient and environmentally friendly operation of the incineration process.

[0003] In one prior art, the control method of solid waste incineration usually relies on preset fixed parameters or a single monitoring index, such as adjusting only according to the furnace temperature, ignoring the real-time differences of solid waste composition and the interactive effects of multiple parameters in the incineration process. This static or single-variable control method is difficult to achieve precise dynamic regulation when facing complex composition or highly variable characteristics of solid waste, often resulting in low combustion efficiency or excessive pollution emissions. Due to the lack of comprehensive analysis and coordinated control of multi-dimensional parameters such as calorific value and moisture content, the system cannot generate an optimal load distribution scheme according to real-time operating conditions, thereby affecting the stability of the equipment and the overall treatment effect.

[0004] In summary, the prior art has the problems of inaccurate incineration process control and low running efficiency. SUMMARY

[0005] The present application provides an automatic control method and system for solid waste incineration equipment to solve the problems of inaccurate incineration process control and low running efficiency.

[0006] In a first aspect, to solve the above technical problems, the present application provides an automatic control method for solid waste incineration equipment, comprising:

[0007] Obtaining real-time data of solid waste composition in the incineration furnace, preprocessing to obtain an initial multi-dimensional parameter set;

[0008] According to the initial multi-dimensional parameter set, fuzzy reasoning is performed to obtain the balance point of calorific value and moisture content;

[0009] If the balance point of calorific value and moisture content exceeds the preset balance point threshold, parameter interaction prediction and analysis are performed to obtain a multi-dimensional control vector;

[0010] According to the multi-dimensional control vector, data fusion is performed to obtain a preliminary scheme framework of load distribution, and historical data matching analysis is performed according to the preliminary scheme framework to obtain an enhanced load distribution adjustment factor;

[0011] According to the enhanced load distribution adjustment factor and the balance point of the calorific value and the moisture content, weight comparison is performed to determine a load balancing influence index, and if the load balancing influence index is greater than a preset compatibility threshold, a matching process is corrected to obtain a corrected matching verification result;

[0012] According to the corrected matching verification result, a load distribution scheme is updated to obtain a final incineration operation strategy;

[0013] According to the final incineration operation strategy, instruction sending and continuous optimization are performed to obtain a closed-loop control response. Preferably, the real-time data of the solid waste composition in the incinerator hearth is obtained, preprocessed to obtain an initial multi-dimensional parameter set, including:

[0014] The real-time data of the solid waste composition in the incinerator hearth is collected from a sensor array to obtain an original data set;

[0015] According to the original data set, cleaning and formatting are performed to obtain cleaned data;

[0016] According to the cleaned data, data fusion is performed to obtain an initial multi-dimensional parameter set.

[0017] Preferably, according to the initial multi-dimensional parameter set, fuzzy reasoning is performed to obtain a balance point of the calorific value and the moisture content, including:

[0018] According to the initial multi-dimensional parameter set, smoothing processing is performed to obtain a smoothed data set;

[0019] According to the smoothed data set, a fuzzy rule set is generated and fuzzy reasoning is performed to obtain a preliminary balance point;

[0020] From the preliminary balance point, a combination parameter is extracted, and if the combination parameter exceeds a preset combination parameter threshold, weight adjustment is performed to obtain a balance point of the calorific value and the moisture content.

[0021] Preferably, if the balance point of the calorific value and the moisture content exceeds a preset balance point threshold, parameter interaction prediction and analysis are performed to obtain a multi-dimensional control vector, including:

[0022] The balance point of the calorific value and the moisture content is compared with the preset balance point threshold, and if it exceeds the preset balance point threshold, parameter interaction prediction is performed to obtain a prediction result;

[0023] Based on the prediction results, the interaction effects of collaborative control are calculated to obtain the adjusted control vector;

[0024] The adjusted control vector is iteratively optimized to obtain a multi-dimensional control vector.

[0025] Preferably, the step of obtaining a preliminary load allocation scheme framework by data fusion based on the multi-dimensional control vector, and obtaining an enhanced load allocation adjustment factor by historical data matching analysis based on the preliminary scheme framework, includes:

[0026] The dynamic operation data of the incineration process is acquired and integrated with the multi-dimensional control vector to obtain a fused dataset;

[0027] If the key indicators in the fused dataset exceed the preset key indicator threshold, then the relationship between the multidimensional indicators is calculated to obtain the dynamically adjusted parameters.

[0028] Based on the aforementioned dynamic adjustment parameters, the control vector is optimized and calculated to obtain the optimized control vector;

[0029] Based on the optimized control vector, load allocation is determined to obtain a preliminary scheme framework;

[0030] Based on the preliminary scheme framework, similar reference cases are matched and the mean of multidimensional indicators is calculated to obtain the enhanced load allocation adjustment factor.

[0031] Preferably, the step of performing a weighted comparison based on the enhanced load distribution adjustment factor and the equilibrium point of calorific value and moisture content to determine the load balancing impact index, and if the load balancing impact index is greater than a preset compatibility threshold, then performing a matching process correction to obtain a corrected matching verification result, includes:

[0032] Based on the dynamic operation data of the incineration process, fuzzy classification is performed and the weight values ​​of each factor are calculated to obtain the weight comparison results;

[0033] If the weight comparison result exceeds the preset weight threshold, the balance point between calorific value and moisture content is adjusted to obtain an optimized balance point.

[0034] Based on the optimized balance point, load balancing metrics are calculated to obtain load balancing impact metrics.

[0035] If the impact index is greater than the preset compatibility threshold, the matching verification result is optimized to obtain the corrected matching verification result.

[0036] Preferably, the step of updating the load allocation scheme based on the corrected matching verification results to obtain the final incineration operation strategy includes:

[0037] Business environment factors are extracted from the dynamic operation data of the incineration process and clustered to obtain the classified load distribution status.

[0038] Based on the classified load distribution status and the corrected matching verification results, the resource utilization and dynamic allocation rules are optimized to obtain an optimized load allocation scheme.

[0039] Based on the optimized load allocation scheme, a decision path is analyzed and generated to obtain the final incineration operation strategy.

[0040] Preferably, the step of sending instructions and continuously optimizing based on the final incineration operation strategy to obtain a closed-loop control response includes:

[0041] Based on the dynamic operating data of the incineration process, time series analysis is performed to obtain the equipment status of the furnace actuator;

[0042] Determine whether the equipment status of the furnace actuator meets the preset operating threshold. If it does not meet the threshold, generate an adjustment instruction set. If it does meet the threshold, generate the instruction set and obtain the instruction sending sequence.

[0043] According to the instruction sending sequence, instructions are sent, and operational feedback data is obtained to obtain a closed-loop control response.

[0044] Secondly, the present invention provides an automatic control device for solid waste incineration equipment, comprising:

[0045] The data acquisition module is used to acquire real-time data on the composition of solid waste in the incinerator, perform preprocessing, and obtain an initial multi-dimensional parameter set.

[0046] The fuzzy inference module is used to perform fuzzy inference based on the initial multi-dimensional parameter set to obtain the balance point between calorific value and moisture content.

[0047] The parameter analysis module is used to perform interactive parameter analysis and obtain a multi-dimensional control vector if the equilibrium point between the calorific value and the moisture content exceeds a preset equilibrium point threshold.

[0048] The fusion and matching module is used to perform data fusion based on the multi-dimensional control vector to obtain a preliminary load allocation scheme framework, and to perform historical data matching analysis based on the preliminary scheme framework to obtain an enhanced load allocation adjustment factor.

[0049] The comparison and correction module is used to perform a weight comparison based on the enhanced load distribution adjustment factor and the balance point of calorific value and moisture content, determine the load balancing impact index, and if the load balancing impact index is greater than the preset compatibility threshold, then the matching process is corrected to obtain the corrected matching verification result.

[0050] The strategy generation module is used to update the load allocation scheme based on the corrected matching verification results to obtain the final incineration operation strategy.

[0051] The closed-loop control module is used to send instructions and continuously optimize based on the final incineration operation strategy to obtain a closed-loop control response.

[0052] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement an automatic control method for a solid waste incineration device as described in any one of the above.

[0053] Fourthly, the present invention also provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform an automatic control method for a solid waste incineration device as described in any one of the above.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] (1) This invention collects multi-dimensional parameters of solid waste in real time and analyzes them in combination with fuzzy logic control and neural network model. It can dynamically adjust the balance point of key parameters according to the real-time dynamic changes of solid waste composition, and accurately predict the interaction between multiple parameters when the working conditions deviate from the norm. This effectively addresses the problem of complex and variable solid waste composition and significantly improves the stability of the incineration process and the ability to adapt to abnormal working conditions.

[0056] (2) This invention matches the preliminary scheme framework generated by real-time analysis with reference cases in the historical database, extracts and verifies the enhanced load allocation adjustment factor, applies the historical best operating experience to the current control decision, so that the final generated incineration operation strategy is more scientific and reliable, optimizes load allocation and improves incineration efficiency.

[0057] (3) By constructing a complete control link from multi-dimensional data acquisition, dynamic analysis, scheme optimization to closed-loop feedback, this invention comprehensively considers multiple key indicators such as calorific value and moisture content and their mutual influence, generates and executes precise control commands, realizes multi-dimensional collaborative control of the entire incineration process, overcomes the limitations of traditional single-parameter control, and ensures the overall operation effect of the incineration process. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of an automatic control method for solid waste incineration equipment provided in the first embodiment of the present invention;

[0059] Figure 2This is a schematic diagram of the structure of an automatic control system for a solid waste incineration equipment provided in the second embodiment of the present invention. Detailed Implementation

[0060] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0061] Reference Figure 1 The first embodiment of the present invention provides an automatic control method for solid waste incineration equipment, comprising the following steps:

[0062] S11: Obtain real-time data on the composition of solid waste in the incinerator, perform preprocessing, and obtain an initial multi-dimensional parameter set;

[0063] S12, based on the initial multi-dimensional parameter set, perform fuzzy reasoning to obtain the balance point between calorific value and moisture content;

[0064] S13, if the equilibrium point between the calorific value and the moisture content exceeds the preset equilibrium point threshold, then parameter interaction prediction and analysis are performed to obtain a multi-dimensional control vector.

[0065] S14. Based on the multi-dimensional control vector, data fusion is performed to obtain a preliminary load allocation scheme framework, and historical data matching analysis is performed based on the preliminary scheme framework to obtain an enhanced load allocation adjustment factor.

[0066] S15. Based on the enhanced load distribution adjustment factor and the balance point of calorific value and moisture content, a weight comparison is performed to determine the load balance impact index. If the load balance impact index is greater than the preset compatibility threshold, the matching process is corrected to obtain the corrected matching verification result.

[0067] S16. Based on the corrected matching verification results, update the load allocation scheme to obtain the final incineration operation strategy;

[0068] S17. Based on the final incineration operation strategy, commands are sent and continuous optimization is performed to obtain a closed-loop control response.

[0069] In step S11, real-time data on the composition of solid waste inside the incinerator is acquired, preprocessed, and an initial multi-dimensional parameter set is obtained, including:

[0070] Real-time data on the composition of solid waste inside the incinerator is collected from a sensor array to obtain the raw dataset;

[0071] Based on the original dataset, the data is cleaned and formatted to obtain the cleaned data;

[0072] Based on the cleaned data, data fusion is performed to obtain an initial multi-dimensional parameter set.

[0073] In step S11, real-time data on the solid waste components are collected in real time by a multi-point sensor array deployed within the incinerator, forming a raw dataset. In one implementation, the sensor array specifically includes an infrared thermal imaging sensor for estimating calorific value by capturing radiation intensity, a humidity sensor for measuring the surface moisture content of solid waste using capacitive or resistive principles, and a gas analyzer for indirectly verifying data accuracy by assisting in the analysis of combustion products. The collected raw dataset is a timestamped data stream containing specific calorific value estimates, such as a calorific value of approximately 5000 kcal / kg corresponding to radiation intensity in a certain area, and moisture content measurements, such as a sample having a moisture content of 30%.

[0074] Subsequently, based on the acquired original dataset, data preprocessing is performed to obtain cleaned data. This preprocessing process includes two stages: cleaning and formatting. The cleaning stage employs median filtering and linear interpolation. Specifically, the median filtering technique involves applying a sliding window of a preset size (e.g., N=5 data points) to the time series data such as calorific value. The data points within the window are numerically sorted, and the original value of the center point of the window is replaced with the sorted median. This process removes outliers caused by high temperatures or signal interference, for example, smoothing invalid calorific value data that jumps to 10000 kcal / kg to a reasonable range of 4000 kcal / kg-6000 kcal / kg. The specific operation of the linear interpolation method is as follows: when a missing data is detected at time point t, the valid data values ​​at the previous time point t-1 and the next time point t+1 are obtained, and the missing value is filled by calculating the arithmetic mean of the two values. For example, the missing value is calculated as 31% based on the moisture content of 30% and 32% at the previous and next time points. The formatting step aligns the data streams from different sensors with timestamps and unifies the sampling frequency, for example, once per second, to obtain the cleaned data.

[0075] Finally, based on the cleaned data, the initial multi-dimensional parameter set is generated using data fusion technology. This fusion process is implemented through a weighted average algorithm, first performing max-min normalization on the cleaned calorific value and moisture content values ​​to linearly map the data to the [0,1] interval. The specific calculation method for normalization is as follows:

[0076] ,in The value is the normalized value, and V is the current value. and These are the minimum and maximum values ​​of the parameter obtained from historical databases. Weighting factors are then set based on their influence on the combustion process. The weighting factors are determined through multiple linear regression analysis of historical operating data. Specifically, a regression model is established using historical combustion efficiency as the dependent variable and normalized calorific value and water content as independent variables. The standardized regression coefficients of each independent variable calculated by the model objectively quantify the influence of each factor on combustion efficiency. Finally, these regression coefficients are normalized (so that their sum is 1) to obtain the weighting factors for each parameter. The calorific value weighting factor W is... h The setting is based on its dominant role in combustion efficiency, with water content weighting factor W. m The setting is based on its impact on combustion stability, and can be set to 0.6 and 0.4 respectively. A comprehensive parameter value is calculated using the following formula: ,in Represents the normalized calorific value. This represents the normalized moisture content. For example, for a solid waste sample with a calorific value of 5500 kcal / kg and a moisture content of 28%, after normalization and weighted calculation, a single comprehensive parameter value will be obtained. This value, together with its corresponding other parameters, constitutes the initial multidimensional parameter set.

[0077] In step S12, fuzzy inference is performed based on the initial multi-dimensional parameter set to obtain the equilibrium point between calorific value and moisture content, including:

[0078] Based on the initial multi-dimensional parameter set, a smoothing process is performed to obtain the smoothed dataset;

[0079] Based on the smoothed dataset, a fuzzy rule set is generated and fuzzy inference is performed to obtain a preliminary equilibrium point;

[0080] Extract the combined parameters from the initial equilibrium point. If the combined parameters exceed the preset combined parameter threshold, then perform weight adjustment to obtain the equilibrium point of calorific value and moisture content.

[0081] In step S12, the initial multi-dimensional parameter set is first smoothed using a moving average filtering algorithm to obtain smoothed data. The specific operation of this smoothing process is as follows: a fixed-size time window is set (e.g., containing 5 data points before and after the current time point, for a total of 11 points). This window is then slid along the time series, and the arithmetic mean of all data points within the window is used as the new value of the window's center point, thereby eliminating instantaneous noise interference in the data stream. For example, if the calorific value data suddenly changes to 9000 kcal / kg due to interference at a certain moment, this method can smooth it to a reasonable range of 5100 kcal / kg.

[0082] Subsequently, based on the smoothed dataset, fuzzification processing is performed to generate a fuzzy rule set, and fuzzy inference is performed based on this rule set to obtain a preliminary equilibrium point. In one implementation, the process first maps the input smoothed calorific value and moisture content parameters to fuzzy linguistic variables such as "high," "medium," and "low." Then, based on historical operating data, a set of fuzzy rule sets in the form of "IF-THEN" is established. The rule set is set based on historical data mining: first, by analyzing a large amount of historical operating data, the statistical correlation between different combinations of calorific value and moisture content ranges and the final combustion efficiency is mined; then, these data correlations are transformed into linguistic control rules. The specific operation of this transformation process is as follows: based on the high-efficiency combustion range determined by statistical analysis (e.g., the range where the calorific value is higher than 5500 kcal / kg and the moisture content is lower than 20%), corresponding fuzzy rules are set, thereby objectively solidifying data insights into control logic. For example, the rule can be: "IF calorific value IS high AND moisture content IS low THEN combustion efficiency IS high." The system fuzzifies the real-time input parameters, matches them against the rule base, performs fuzzy inference operations, and outputs a quantified result, which is the preliminary equilibrium point, containing parameters such as preliminary calorific value and preliminary moisture content.

[0083] Finally, combined parameters (specifically, preliminary calorific value parameters) are extracted from the preliminary equilibrium point, and it is determined whether they exceed a preset combined parameter threshold (specifically, a preset calorific value threshold). If they do, a weighted average algorithm is used for weight adjustment to obtain the final equilibrium point between calorific value and moisture content. The preset calorific value threshold is set based on the design specifications of the incinerator and statistical analysis of historical data from long-term stable operation; for example, it can be set to 6000 kcal / kg. When the preliminary calorific value parameter (e.g., 6200 kcal / kg) exceeds this threshold, weight adjustment is initiated. The specific calculation for this adjustment process is as follows: First, an adjustment coefficient is calculated based on the preliminary moisture content, and then this coefficient is used to correct the preliminary calorific value. This adjustment process aims to weight and smooth the excess parameter values ​​towards their threshold to achieve a stable transition. The specific calculation formula is as follows:

[0084] ,in This represents the adjusted calorific value. Represents preliminary calorific value. This represents a preset calorific value threshold, which is set based on the design specifications of the incinerator and statistical analysis of historical data from long-term stable operation. For example, it can be set to 6000 kcal / kg, W. h and W mThese are preset weighting factors for calorific value and moisture content, set based on the degree of influence of each parameter on combustion stability; for example, they can be set to 0.7 and 0.3 respectively. Based on this calculation, an excessively high initial calorific value is adjusted to a more optimal value. For example, for an initial calorific value of 6200 kcal / kg, the adjusted calorific value is (6200 × 0.7) + (6000 × 0.3) = 4340 + 1800 = 6140 kcal / kg. This adjusted calorific value, together with the corresponding moisture content, constitutes the equilibrium point between calorific value and moisture content. If the initial calorific value parameter does not exceed the threshold, the initial equilibrium point is directly used as the equilibrium point between calorific value and moisture content.

[0085] In step S13, if the equilibrium point between calorific value and moisture content exceeds a preset equilibrium point threshold, parameter interaction prediction and analysis are performed to obtain a multi-dimensional control vector, including:

[0086] The equilibrium point of calorific value and moisture content is compared with a preset equilibrium point threshold. If it exceeds the preset equilibrium point threshold, parameter interaction prediction is performed to obtain the prediction result.

[0087] Based on the prediction results, the interaction effects of collaborative control are calculated to obtain the adjusted control vector;

[0088] The adjusted control vector is iteratively optimized to obtain a multi-dimensional control vector.

[0089] First, the parameters (such as calorific value, moisture content, and particle size) at the equilibrium point of calorific value and moisture content are compared with preset equilibrium point thresholds. The preset equilibrium point threshold is a multi-dimensional parameter range, officially called the incinerator stable operation parameter threshold. This threshold is set based on the equipment's design specifications and historical data statistics ensuring safe and stable combustion. For example, this threshold can be specifically set as a calorific value not exceeding 6000 kcal / kg and a moisture content not exceeding 30%. If any parameter at the current equilibrium point exceeds this threshold range, an interactive prediction between parameters is initiated to obtain the prediction result.

[0090] The prediction process is achieved through a pre-built and trained neural network prediction model. The model construction process is as follows: First, a large amount of historical operating data from the incinerator is collected to construct a training dataset containing multi-dimensional inputs and a single output. The input features are combinations of parameters such as historical calorific value, moisture content, and particle size, while the output label is the actual incineration efficiency corresponding to each input combination. Then, the network structure and hyperparameters are optimized and determined through a combination of grid search and k-fold cross-validation. The specific steps are as follows: The search range for hyperparameters is preset, for example, the number of hidden layers ranges from [1,5], the number of neurons per layer ranges from [10,100], and the learning rate ranges from [0.001,0.1]. The training dataset is divided into k (e.g., k=10) mutually exclusive subsets of similar size. The model with different hyperparameter combinations is trained sequentially using k-1 subsets, and validated using the remaining subset, repeated k times. Finally, the set of hyperparameters with the smallest average error in the k validations (e.g., determined to be 2 hidden layers with 64 neurons per layer) is selected as the optimal configuration. The training process employs the backpropagation algorithm, which continuously adjusts the connection weights and biases between neurons within the network to minimize the root mean square error function. The specific form of this error function is as follows: Where n is the number of samples, Let i be the actual incineration efficiency of the i-th sample. To assess the model's prediction efficiency for this sample, once the root mean square error of the model's predictions on an independent validation dataset is less than a preset prediction accuracy threshold (e.g., 5%), its network structure and weight parameters are saved to form the final prediction model. The preset prediction accuracy threshold is set based on ensuring the model's generalization ability and avoiding overfitting while meeting the minimum accuracy requirements for subsequent control decisions. This threshold is determined through multiple trials on the test set. The specific operation for parameter interaction prediction involves feeding the current real-time parameter set exceeding the threshold (e.g., calorific value 5200 kcal / kg, moisture content 28%, particle size 2.5 mm) as input data into the input layer of the trained neural network model. Data propagates forward through the network. In each layer, the input data is linearly multiplied and added to the layer's weight matrix, and a bias vector is added. For a layer with m inputs and n neurons, its weight matrix is ​​an m×n matrix. Subsequently, the calculation result is transformed using a non-linear activation function (such as the ReLU function) and passed to the next layer. The process of performing linear transformations and nonlinear activations layer by layer constitutes a nonlinear analysis of the dynamic interaction effects of multidimensional parameters. Ultimately, the network output layer obtains a quantitative prediction result, which is specifically a rating of the incineration efficiency under the current operating conditions, such as "moderately high".

[0091] Next, based on the predicted results, a collaborative control interaction effect calculation is performed to obtain the adjusted control vector. This calculation uses a weighted average algorithm, assigning preset weight factors to each input parameter (calorific value, moisture content, particle size). The weight factors are set based on the primary and secondary relationships of each parameter's influence on the incineration process; for example, a weight W can be assigned to the calorific value. h =0.6, assigning a weight W to the moisture content. m =0.3, assigning a weight W to the particle size. ps =0.1. The calculation process generates a set of corrected target parameter values, which constitute the adjusted control vector. For example, for an input with a calorific value of 5800 kcal / kg, the calculation yields an optimization result that adjusts the target calorific value to 5600 kcal / kg.

[0092] Finally, the adjusted control vector is iteratively optimized through a real-time processing module to obtain the final multi-dimensional control vector. This iterative optimization is implemented as a feedback adjustment loop. The system dynamically adjusts the actual operating parameters of the incinerator, such as the feed rate and oxygen supply, based on the adjusted control vector as the target. For example, if the predicted incineration efficiency rating is lower than a preset efficiency threshold (e.g., below 70 points out of 100), it is determined to be inefficient. The system instructs to reduce the feed rate to 0.8 t / h and simultaneously increase the oxygen supply to 1200 m³ / h. The system monitors the operating parameters from the feed system speed sensor and the pipeline oxygen flow meter in real time and continuously compares them with the optimization target (feed rate 0.8 t / h, oxygen supply 1200 m³ / h) until the monitored operating parameters stabilize near the optimization target. The loop ends at this point, and the stabilized set of actual operating parameters constitutes the multi-dimensional control vector.

[0093] In step S14, based on the multi-dimensional control vector, data fusion is performed to obtain a preliminary load allocation scheme framework, and historical data matching analysis is performed based on the preliminary scheme framework to obtain an enhanced load allocation adjustment factor, including:

[0094] The dynamic operation data of the incineration process is acquired and integrated with the multi-dimensional control vector to obtain a fused dataset;

[0095] If the key indicators in the fused dataset exceed the preset key indicator threshold, then the relationship between the multidimensional indicators is calculated to obtain the dynamically adjusted parameters.

[0096] Based on the aforementioned dynamic adjustment parameters, the control vector is optimized and calculated to obtain the optimized control vector;

[0097] Based on the optimized control vector, load allocation is determined to obtain a preliminary scheme framework;

[0098] Based on the preliminary scheme framework, similar reference cases are matched and the mean of multidimensional indicators is calculated to obtain the enhanced load allocation adjustment factor.

[0099] First, dynamic operational data of the incineration process is acquired. This data is a real-time data stream containing multi-dimensional indicators such as calorific value, moisture content, and particle size. Data fusion technology is then used to integrate this data with the multi-dimensional control vectors to obtain a fused dataset. The specific implementation of this data fusion technology employs a Kalman filter algorithm. In one implementation, the operation includes two stages: prediction and update. In the prediction stage, the algorithm predicts the current state based on the state estimate from the previous moment and the process noise covariance. In the update stage, the algorithm combines the actual observations at the current moment (i.e., multi-source data from sensors) and calculates the Kalman gain to weight and correct the predicted value, obtaining an optimal estimate closer to the true value. This results in a fused dataset with less data fluctuation and higher consistency.

[0100] Next, it is determined whether the key indicators (specifically calorific value, moisture content, and particle size) in the fused dataset exceed preset key indicator thresholds. These preset key indicator thresholds are set based on the optimal operating range to ensure incineration efficiency and equipment safety, derived through historical big data analysis. For example, they can be set to a calorific value not exceeding 5500 kcal / kg and a moisture content not exceeding 28%. If these thresholds are exceeded, a multi-dimensional indicator relationship calculation is initiated to obtain dynamically adjusted parameters. The specific process of relation calculation involves using principal component analysis. First, Z-score standardization is performed on multiple indicators in the fused dataset, such as calorific value and moisture content. This involves subtracting the mean from the data for each indicator and then dividing by its standard deviation to eliminate the influence of dimensions. Next, the covariance matrix of the standardized data is calculated, and eigenvalues ​​and eigenvectors are obtained through eigenvalue decomposition. Finally, the eigenvector corresponding to the largest eigenvalue is selected as the first principal component. The coefficients of this principal component reveal the strength and direction of the intrinsic correlation between the indicators. By calculating the ratio of the loading coefficients of two original indicators (e.g., calorific value and moisture content) in this principal component and multiplying this ratio by the ratio of the standard deviations of the two original indicators, the strength and direction of the intrinsic correlation between the indicators are quantified. This quantified result is the dynamic adjustment parameter. For example, it is calculated that for every 1% increase in moisture content, the calorific value decreases by approximately 50 kcal / kg. Then, based on the dynamic adjustment parameter, the control vector is optimized using a weighted average algorithm to obtain the optimized control vector.

[0101] Subsequently, based on the optimized control vector, load allocation is determined to obtain a preliminary scheme framework. The specific operation and process of load allocation determination are as follows: First, a historical database is constructed. This database is a structured dataset that continuously records and stores operating parameters (calorific value, moisture content, etc.), control commands (load allocation ratio), and operating results (combustion efficiency, emission level) during the long-term operation of the incinerator. Then, the historical database is analyzed offline to generate a tree-like decision structure. At each decision node, this generation process systematically evaluates all possible split points for each operating parameter and selects the split point that makes the divided data subset most consistent in terms of load allocation scheme as the optimal judgment condition (e.g., "calorific value > 5000 kcal / kg"). This process is executed recursively until a preset stopping condition is met, ultimately forming a decision structure containing the optimal decision path, where each leaf node stores a specific load allocation scheme (e.g., "main combustion chamber 70%, auxiliary combustion chamber 30%"). The real-time operation for load allocation judgment is as follows: the parameter values ​​in the optimized control vector are input from the root node of the decision tree, and the tree is traversed layer by layer downwards according to the judgment conditions set by the node until a leaf node is reached. The load allocation scheme stored in the leaf node is then output as the preliminary scheme framework.

[0102] Finally, based on the preliminary scheme framework, similar reference cases are matched in the historical database, and their multidimensional index mean values ​​are calculated to obtain the enhanced load allocation adjustment factor. This process first classifies the cases in the historical database using a K-means clustering algorithm, forming multiple operating condition categories. The number of clusters, K, is determined by calculating the sum of squared errors within clusters (WCSS) at different K values ​​(e.g., from 2 to 10), plotting the K-WCSS relationship curve, and selecting the K value corresponding to the elbow point where the rate of decline in the curve decreases from rapid to slow as the optimal number of clusters. Then, based on the characteristics of the preliminary scheme framework, one or more cases with similar historical operating characteristics are retrieved and matched from the classified case set. Finally, a weighted average is calculated for all matched similar cases' multidimensional indexes (such as calorific value and moisture content). This calculation result is the enhanced load allocation adjustment factor, which is a set of parameter values ​​validated by historical data and closer to the optimal operating condition.

[0103] In step S15, a weighted comparison is performed based on the enhanced load distribution adjustment factor and the equilibrium point of calorific value and moisture content to determine the load balancing impact index. If the load balancing impact index is greater than a preset compatibility threshold, a matching process correction is performed to obtain the corrected matching verification result, including:

[0104] Based on the dynamic operation data of the incineration process, fuzzy classification is performed and the weight values ​​of each factor are calculated to obtain the weight comparison results;

[0105] If the weight comparison result exceeds the preset weight threshold, the balance point between calorific value and moisture content is adjusted to obtain an optimized balance point.

[0106] Based on the optimized balance point, load balancing metrics are calculated to obtain load balancing impact metrics.

[0107] If the impact index is greater than the preset compatibility threshold, the matching verification result is optimized to obtain the corrected matching verification result.

[0108] First, based on the dynamic operational data of the incineration process (specifically, a set of operational environmental factors including key parameters such as waste calorific value, moisture content, oxygen supply, and feed rate), fuzzy classification is performed and the weight values ​​of each factor are calculated to obtain weight comparison results. The specific transformation rule for this fuzzy classification operation is to pre-define a triangular membership function for each factor's fuzzy linguistic variables such as "high," "medium," and "low." This function is defined by three points (left boundary, center point, and right boundary) within a parameter's value range. For any input continuous value (e.g., calorific value 4800 kcal / kg), its membership degree to each fuzzy linguistic variable is determined by calculating its vertical position on this triangle (e.g., "medium calorific value" membership degree 0.7, "high calorific value" membership degree 0.3). Subsequently, multiple linear regression analysis is performed on historical operational data to objectively calculate the weight values ​​of each factor. The specific steps of this process are as follows: First, extract a dataset from the historical database containing various business environment factors (independent variables) and corresponding process stability indicators (dependent variables, such as the standard deviation of furnace temperature). Then, establish a multiple linear regression model based on this dataset and calculate the standardized regression coefficient for each factor. The absolute value of this coefficient objectively quantifies the degree of influence of each factor on process stability. Finally, normalize the absolute values ​​of the standardized regression coefficients of each factor to obtain the final weight value. The calculation formula is: the weight value of a factor is equal to the absolute value of the factor's regression coefficient divided by the sum of the absolute values ​​of all factor regression coefficients. After calculating the weight values ​​of all factors, the largest weight value and its corresponding factor are taken as the weight comparison result. Next, determine whether the weight comparison result exceeds a preset weight threshold. The preset weight threshold, also known as the key influencing factor weight determination threshold, is set based on statistical analysis of historical data to determine the minimum weight value that has a significant impact on the stability of the incineration process; for example, it can be set to 0.45. If the weight comparison result (e.g., calorific value weight 0.5) exceeds the threshold, the balance point between calorific value and moisture content is adjusted to obtain an optimized balance point. In this context, the balance point between calorific value and moisture content refers to a set of target control parameters determined by the preceding step (S12) to guide the current combustion process. The adjustment operation compares the current real-time data (e.g., calorific value 4850 kcal / kg) with the average parameter values ​​of similar operating conditions in the historical database, calculates the deviation between the two, and adjusts the current control parameters according to the deviation, such as adjusting the oxygen supply to 1020 m³ / h and the feed rate to 0.9 t / h. This set of adjusted parameters is the optimized balance point.

[0109] Then, based on the optimized equilibrium point, load balancing indices are calculated using preset decision rules to obtain load balancing impact indices. This calculation process takes the parameter values ​​from the optimized equilibrium point as input and matches them with a set of IF-THEN format decision rules, such as: "IF calorific value < 4900 kcal / kg THEN main combustion chamber load = 60%"; "IF moisture content > 25% THEN start pre-drying". A set of specific operating instructions or status adjustment values ​​output after matching the rules (e.g., oxygen supply adjusted to 980 m³ / h, feed rate reduced to 0.85 t / h) constitutes the load balancing impact indices.

[0110] Finally, it is determined whether the coordination matching degree between the influencing index and the current control strategy is greater than a preset compatibility threshold. This preset compatibility threshold, also known as the historical and real-time strategy compatibility threshold, is set based on the minimum matching degree required to ensure that introducing historical experience will not disturb the current real-time control system; for example, it can be set to 0.8. If the coordination matching degree is greater than this threshold, the matching verification result is optimized to obtain a corrected matching verification result. The specific process of this optimization operation is as follows: First, the current real-time operating parameters (e.g., real-time calorific value 4850 kcal / kg, moisture content 26%) are used as a query vector; then, the similarity between this query vector and the central feature vectors of each operating condition category (e.g., "medium calorific value, medium moisture content" category) pre-divided through cluster analysis in the historical database is calculated. The specific process of pre-dividing through cluster analysis is as follows: First, by calculating the sum of squared errors within clusters under different K values ​​and plotting curves, the K value corresponding to the "elbow point" of the curve is selected to determine the optimal number of clusters, K value. Then, using a defined K value as a parameter, K-means clustering is performed on all operating condition data in the historical database, dividing the historical data into K operating condition categories, and calculating the central feature vector for each category. The specific method for similarity calculation is to calculate the Euclidean distance between two vectors; a smaller distance value indicates higher similarity. This calculation identifies and determines the best-matching historical operating condition category. Next, the system extracts pre-stored historical optimal operating control parameters associated with this category (e.g., optimal oxygen supply of 990 m³ / h and optimal feed rate of 0.88 t / h). Finally, the system directly uses this set of extracted historical optimal control parameters as the final command parameters for the current control cycle, updating the current control parameter setpoints. This set of parameters updated with historical optimal values ​​constitutes the corrected matching verification result.

[0111] In step S16, based on the corrected matching verification results, the load allocation scheme is updated to obtain the final incineration operation strategy, including:

[0112] Business environment factors are extracted from the dynamic operation data of the incineration process and clustered to obtain the classified load distribution status.

[0113] Based on the classified load distribution status and the corrected matching verification results, the resource utilization and dynamic allocation rules are optimized to obtain an optimized load allocation scheme.

[0114] Based on the optimized load allocation scheme, a decision path is analyzed and generated to obtain the final incineration operation strategy.

[0115] First, operational environmental factors are extracted from the dynamic operational data of the incineration process. These factors specifically include operational status data such as incinerator temperature, waste feed rate, and flue gas emission concentration (e.g., CO content). Next, the K-means clustering algorithm is used to classify the operational status data, resulting in a classified load distribution. The optimal number of clusters, K, is determined by calculating the sum of squared errors within clusters at different K values ​​and plotting a curve. The K value corresponding to the inflection point of the curve is selected as the optimal number of clusters. Specifically, the clustering algorithm treats the operational status data at each time point (e.g., temperature 900°C, feed rate 1.0 t / h, CO content 80 ppm) as a data point in a multi-dimensional space. The algorithm divides these data points into several clusters based on their spatial distance. The center point of each cluster represents a typical load distribution state, such as "high temperature, high feed rate" or "medium temperature, low emission."

[0116] Then, based on the classified load distribution status and the corrected matching verification results obtained in the previous step, the resource utilization and dynamic allocation rules are optimized through a combination of parameter matching and rule adjustment to obtain an optimized load allocation scheme. The specific calculation of this optimization process is as follows: First, the current real-time operating status is matched to the closest classified load distribution status (e.g., the "high temperature, high feed" category); next, the system retrieves the baseline resource allocation rule associated with this status from the configuration library (e.g., the baseline air volume is 1000 m³ / h); finally, the system uses the control parameters verified by historical data in the corrected matching verification results to adjust the baseline rule. For example, if the optimal air volume in the verification results is 1100 m³ / h, this value is used as the final adjustment result, thereby generating an optimized load allocation scheme, such as adjusting the main combustion chamber air volume to 1100 m³ / h.

[0117] Finally, based on the optimized load allocation scheme, a set of preset decision rules is executed to analyze and generate decision paths, resulting in the final incineration operation strategy. The specific operation of generating this decision path involves taking the parameters from the optimized load allocation scheme as input and sequentially traversing and judging them within a preset "IF-THEN" decision rule base. This decision rule base is based on historical data analysis and safety procedure settings and is used to handle specific operating conditions. For example, one rule is: "IF CO content > 90ppm, THEN increase oxygen supply to 1050m³ / h AND reduce feed rate to 0.85t / h." Another rule is: "IF temperature continuously > 930°C, THEN optimize main combustion chamber airflow ratio to 60%." The system analyzes along the matched rule paths, integrates the operation instructions corresponding to all triggered rules, and forms a final, directly executable set of control instructions, which is the final incineration operation strategy.

[0118] In step S17, according to the final incineration operation strategy, instructions are sent and continuous optimization is performed to obtain a closed-loop control response, including:

[0119] Based on the dynamic operating data of the incineration process, time series analysis is performed to obtain the equipment status of the furnace actuator;

[0120] Determine whether the equipment status of the furnace actuator meets the preset operating threshold. If it does not meet the threshold, generate an adjustment instruction set. If it does meet the threshold, generate the instruction set and obtain the instruction sending sequence.

[0121] According to the instruction sending sequence, instructions are sent, and operational feedback data is obtained to obtain a closed-loop control response.

[0122] First, based on the dynamic operating data of the incineration process, specifically the real-time operating data of furnace actuators (such as dampers and oxygen supply fans) collected from sensors (e.g., damper opening, oxygen supply, combustion chamber pressure), a time series analysis method is used to process the data to obtain the equipment status of the furnace actuators. The specific operation of this time series analysis involves using a sliding window of a preset size (e.g., 10 seconds) to decompose the continuous operating data of the actuators into trend and periodic characteristics, thereby generating an equipment status profile representing its current operating state, such as "stable operation" or "high-frequency adjustment."

[0123] Next, it is determined whether the equipment status of the furnace actuator meets a preset operating threshold. This preset operating threshold, also known as the actuator stable operating condition threshold, is set based on the actuator's mechanical performance specifications and the parameter requirements for ensuring stable combustion processes. For example, it can be set as a damper opening of 45%-55% and a combustion chamber pressure within the range of -40 to -30 Pa. The specific operation of this determination process is as follows: First, the analyzed equipment status parameters (e.g., damper opening 48%, pressure -35 Pa) are input into a preset probability calculation function. This function statistically fits the relationship between the equipment status parameters and the probability of normal equipment operation based on historical data, thereby outputting a quantified probability value. Then, the calculated probability (e.g., 0.92) is compared with a preset probability judgment threshold (e.g., 0.8). The probability judgment threshold is set based on optimizing the decision critical point through statistical analysis of historical data. The analysis process plotted a curve showing the relationship between the proportion of the system correctly identifying "abnormal states" and the proportion of incorrectly identifying "normal states" as "abnormal states" at different probability thresholds (i.e., the receiver operating characteristic ROC curve). The optimal critical point is the probability value on this curve that achieves the best balance between the two ratios while meeting preset business requirements; this value is selected as the preset probability judgment threshold. If the calculated real-time probability is higher than this threshold, the equipment state is determined to meet the requirements for stable operation, and the system generates a set of instructions to maintain the current state (e.g., maintain oxygen supply at 1000 m³ / h) according to the final incineration operation strategy. If the calculated probability is lower than this probability judgment threshold, the equipment state is determined to not meet the threshold, and the system generates a set of adjustment instructions for correction (e.g., adjust oxygen supply to 1100 m³ / h, reduce damper opening to 40%). The generated instruction set is the instruction transmission sequence.

[0124] Finally, according to the command sequence, the command set is sent through a feedback loop mechanism, and operational feedback data is acquired to obtain a closed-loop control response. This process transmits the command sequence to the corresponding furnace actuator to activate it. Simultaneously, sensors continuously monitor the actuator's operational feedback data (e.g., the actual pressure value after activation). If the feedback data shows that the actuator state has recovered to within the preset operating threshold (e.g., pressure recovered to -30Pa), the current closed-loop control is complete. If the feedback data shows that the state is still not up to standard (e.g., pressure is still -15Pa), the system uses this feedback data as input for a new round of control, iteratively generating a new adjustment command set and sending it again until the actuator state meets the standard, thus achieving a complete closed-loop control response.

[0125] In summary, this invention achieves precise, stable, and efficient closed-loop automatic control of the solid waste incineration process with dynamic changes in composition by real-time collection of multi-dimensional characteristic parameters of solid waste, combined with fuzzy logic and neural network collaborative prediction optimization, and by introducing a historical case database to correct and optimize the control scheme.

[0126] Reference Figure 2 The second embodiment of the present invention provides an automatic control system for a solid waste incineration equipment, comprising:

[0127] The data acquisition module is used to acquire real-time data on the composition of solid waste in the incinerator, perform preprocessing, and obtain an initial multi-dimensional parameter set.

[0128] The fuzzy inference module is used to perform fuzzy inference based on the initial multi-dimensional parameter set to obtain the balance point between calorific value and moisture content.

[0129] The parameter analysis module is used to perform interactive parameter analysis and obtain a multi-dimensional control vector if the equilibrium point between the calorific value and the moisture content exceeds a preset equilibrium point threshold.

[0130] The fusion and matching module is used to perform data fusion based on the multi-dimensional control vector to obtain a preliminary load allocation scheme framework, and to perform historical data matching analysis based on the preliminary scheme framework to obtain an enhanced load allocation adjustment factor.

[0131] The comparison and correction module is used to perform a weight comparison based on the enhanced load distribution adjustment factor and the balance point of calorific value and moisture content, determine the load balancing impact index, and if the load balancing impact index is greater than the preset compatibility threshold, then the matching process is corrected to obtain the corrected matching verification result.

[0132] The strategy generation module is used to update the load allocation scheme based on the corrected matching verification results to obtain the final incineration operation strategy.

[0133] The closed-loop control module is used to send instructions and continuously optimize based on the final incineration operation strategy to obtain a closed-loop control response.

[0134] It should be noted that the automatic control system for solid waste incineration equipment provided in this embodiment of the invention is used to execute all the process steps of the automatic control method for solid waste incineration equipment in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0135] This invention also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as an automatic control program for solid waste incineration equipment. When the processor executes the computer program, it implements the steps described in the various embodiments of the automatic control method for solid waste incineration equipment, for example... Figure 1 The step S11 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above-described device embodiments, such as the fuzzy inference module.

[0136] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0137] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0138] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting various parts of the electronic device through various interfaces and lines.

[0139] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0140] Wherein, if the modules / units integrated in the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0141] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0142] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for automatically controlling a solid waste incineration facility, characterized by, The method comprises the following steps: acquiring real-time data of solid waste components in the incinerator hearth, preprocessing the data, and obtaining an initial multi-dimensional parameter set; performing fuzzy reasoning based on the initial multi-dimensional parameter set to obtain a balance point of the calorific value and the moisture content; if the balance point of the calorific value and the moisture content exceeds a preset balance point threshold, performing parameter interaction prediction and analysis to obtain a multi-dimensional control vector; performing data fusion based on the multi-dimensional control vector to obtain a preliminary scheme framework of load distribution, and performing historical data matching analysis based on the preliminary scheme framework to obtain an enhanced load distribution adjustment factor; performing weight comparison based on the enhanced load distribution adjustment factor and the balance point of the calorific value and the moisture content to determine a load balancing influence index, and if the load balancing influence index is greater than a preset compatibility threshold, performing matching process correction to obtain a corrected matching verification result; performing load distribution scheme updating based on the corrected matching verification result to obtain a final incineration operation strategy; performing instruction sending and continuous optimization based on the final incineration operation strategy to obtain a closed-loop control response; wherein a comprehensive parameter value is calculated by the following formula: wherein represents a normalized heating value, represents a normalized moisture content, the comprehensive parameter value and the heating value and the moisture content corresponding thereto jointly constitute the initial multi-dimensional parameter set; wherein, if the balance point of the calorific value and the moisture content exceeds a preset balance point threshold, performing parameter interaction prediction and analysis to obtain a multi-dimensional control vector, comprising: comparing the balance point of the calorific value and the moisture content with the preset balance point threshold, and if the balance point exceeds the preset balance point threshold, performing parameter interaction prediction to obtain a prediction result; performing collaborative control interaction influence calculation based on the prediction result to obtain an adjusted control vector; iteratively optimizing the adjusted control vector to obtain a multi-dimensional control vector; wherein, the iterative optimization operation specifically comprises: dynamically adjusting the actual operation parameters of the incinerator, including the feeding speed and the oxygen supply amount, based on the adjusted control vector as the target; continuously comparing the monitored operation parameters from the feeding system speed sensor and the pipeline oxygen flow meter with the optimization target until the monitored operation parameters are stabilized near the optimization target, at which time the stabilized actual operation parameter set constitutes the multi-dimensional control vector.

2. The automatic control method of the solid waste incineration facility according to claim 1, characterized by, The method comprises the following steps: collecting real-time data of solid waste components in the incinerator hearth from a sensor array to obtain an original data set; performing cleaning and formatting based on the original data set to obtain cleaned data; performing data fusion based on the cleaned data to obtain an initial multi-dimensional parameter set.

3. The automatic control method of the solid waste incineration facility according to claim 1, characterized by, The method comprises the following steps: performing smoothing processing based on the initial multi-dimensional parameter set to obtain a smoothed data set; generating a fuzzy rule set based on the smoothed data set and performing fuzzy reasoning to obtain a preliminary balance point; extracting a combination parameter from the preliminary balance point, and if the combination parameter exceeds a preset combination parameter threshold, performing weight adjustment to obtain a balance point of the calorific value and the moisture content.

4. The automatic control method of the solid waste incineration facility according to claim 1, characterized by, The data fusion is performed according to the multi-dimensional control vector to obtain a preliminary scheme framework of load distribution, and historical data matching analysis is performed according to the preliminary scheme framework to obtain an enhanced load distribution adjustment factor, including: Obtain dynamic operation data of the incineration process and integrate with the multi-dimensional control vector to obtain a fusion data set; If the key indicators in the fusion data set exceed the preset key indicator threshold, perform multi-dimensional index relationship calculation to obtain a dynamic adjustment parameter; According to the dynamic adjustment parameter, the control vector is optimized to obtain an optimized control vector; According to the optimized control vector, load distribution judgment is performed to obtain a preliminary scheme framework; According to the preliminary scheme framework, similar reference case matching and multi-dimensional index mean calculation are performed to obtain an enhanced load distribution adjustment factor.

5. The automatic control method of the solid waste incineration facility according to claim 4, characterized by, According to the enhanced load distribution adjustment factor and the balance point of the heat value and the moisture content, weight comparison is performed to determine the load balancing impact indicator, and if the load balancing impact indicator is greater than the preset compatibility threshold, the matching process is corrected to obtain a corrected matching verification result, including: According to the dynamic operation data of the incineration process, fuzzy classification and calculation of each factor weight value are performed to obtain a weight comparison result; If the weight comparison result exceeds the preset weight threshold, the balance point of the heat value and the moisture content is adjusted to obtain an optimized balance point; According to the optimized balance point, load balancing index calculation is performed to obtain a load balancing impact indicator; If the impact indicator is greater than the preset compatibility threshold, the matching verification result is optimized to obtain a corrected matching verification result.

6. The automatic control method of the solid waste incineration facility according to claim 5, characterized by, According to the corrected matching verification result, the load distribution scheme is updated to obtain a final incineration operation strategy, including: Extract business environment factors from the dynamic operation data of the incineration process and perform clustering to obtain a classified load distribution state; According to the classified load distribution state and the corrected matching verification result, the resource utilization rate and the dynamic allocation rule are optimized to obtain an optimized load distribution scheme; According to the optimized load distribution scheme, the decision path is analyzed and generated to obtain a final incineration operation strategy.

7. The automatic control method of the solid waste incineration facility according to claim 6, characterized by, According to the final incineration operation strategy, instruction sending and continuous optimization are performed to obtain a closed-loop control response, including: According to the dynamic operation data of the incineration process, time series analysis is performed to obtain the device state of the furnace executor; Determine whether the device state of the furnace executor meets the preset operation threshold, if not, generate an adjustment instruction set, if yes, perform instruction set generation to obtain an instruction sending sequence; According to the instruction sending sequence, instruction sending is performed, and operation feedback data is obtained to obtain a closed-loop control response.

8. An automatic control system for solid waste incineration equipment, characterized in that, A solid waste incineration equipment automatic control method for realizing the method according to any one of claims 1 to 7, comprising: A data acquisition module for acquiring real-time data of solid waste components in an incinerator hearth, preprocessing to obtain an initial multi-dimensional parameter set; A fuzzy reasoning module for performing fuzzy reasoning according to the initial multi-dimensional parameter set to obtain a balance point of heat value and moisture content; The parameter analysis module is configured to perform parameter interaction analysis to obtain a multi-dimensional control vector if the balance point of the heat value and the moisture content exceeds a preset balance point threshold. The fusion and matching module is configured to perform data fusion to obtain a preliminary scheme framework of load distribution according to the multi-dimensional control vector, and perform historical data matching analysis according to the preliminary scheme framework to obtain an enhanced load distribution adjustment factor. The comparison and correction module is configured to perform weight comparison according to the enhanced load distribution adjustment factor and the balance point of the heat value and the moisture content to determine a load balancing impact index, and perform matching process correction to obtain a corrected matching verification result if the load balancing impact index is greater than a preset compatibility threshold. The strategy generation module is configured to perform load distribution scheme updating according to the corrected matching verification result to obtain a final incineration operation strategy. The closed-loop control module is configured to perform instruction sending and continuous optimization according to the final incineration operation strategy to obtain a closed-loop control response.

Images