Thermal power generating unit intelligent operation auxiliary decision-making system based on big data analysis

Abstract

The invention relates to the technical field of thermal power generating unit intelligent operation, in particular to a thermal power generating unit intelligent operation auxiliary decision-making system based on big data analysis, which comprises a water cooling wall working condition synchronous acquisition module, a temperature difference dynamic compensation module, a soot deposition rate analysis module, a soot blowing demand prediction module and an intelligent soot blowing decision-making module. Wherein the temperature difference dynamic compensation module is used for executing humidity compensation calculation on pipe wall temperature difference data and outputting compensated temperature difference data; the ash deposition rate analysis module is used for calculating current water-cooled wall ash deposition rate data; the soot blowing demand prediction module is used for calculating remaining effective time data for reaching the maximum allowable soot deposition thickness; the intelligent soot blowing decision module is used for matching a corresponding soot blowing strategy; according to the method, through multi-source data fusion and dynamic soot blowing strategy matching, accurate prediction and efficient soot blowing decision making of water-cooled wall soot deposition are achieved, and the safety and economical efficiency of the boiler are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent operation of thermal power generating sets, and in particular to an intelligent operation auxiliary decision-making system for thermal power generating sets based on big data analysis. Background Art

[0002] During the operation of large-scale thermal power units, the stability of the water-cooled wall is of great significance to the safety and efficiency of the boiler. A large amount of ash easily adheres to the surface of the water-cooled wall tubes, resulting in a decrease in heat transfer performance and possibly aggravated damage to the tube walls in local high-temperature areas. The traditional timed sootblowing mode cannot accurately identify the degree and duration of ash accumulation in different areas, resulting in frequent excessive or insufficient sootblowing, which limits the overall thermal economy and safety of the boiler.

[0003] Existing technologies lack comprehensive evaluation methods for key factors such as real-time changes in flue gas flow rate, tube wall temperature difference, and ash melting point, and are unable to efficiently balance ash accumulation with boiler load changes. To meet the operating requirements of large-scale thermal power units under complex coal-fired conditions, it is necessary to achieve dynamic fusion of multi-source monitoring data through big data analysis methods, so as to accurately obtain the ash accumulation rate while timely evaluating the soot blowing demand and making intelligent decisions. Summary of the Invention

[0004] Based on the above objectives, the present invention provides an intelligent operation auxiliary decision-making system for thermal power units based on big data analysis.

[0005] An intelligent operation auxiliary decision-making system for thermal power units based on big data analysis includes a water-wall operating condition synchronous acquisition module, a temperature difference dynamic compensation module, a soot accumulation rate analysis module, a soot blowing demand prediction module, and an intelligent soot blowing decision-making module; wherein:

[0006] Water-wall operating condition synchronous acquisition module: It is used to collect the temperature difference data of each panel tube wall in real time through the distributed temperature sensor array arranged on each panel of the boiler water-wall, and synchronously obtain the ash content data of the coal entering the furnace and the flue gas flow rate data at the corresponding time;

[0007] Temperature difference dynamic compensation module: used to receive tube wall temperature difference data, ash content data and flue gas flow rate data, and perform humidity compensation calculation on the tube wall temperature difference data based on the real-time humidity data obtained by the ambient humidity sensor, and output the compensated temperature difference data;

[0008] Ash accumulation rate analysis module: used to input the compensated temperature difference data into the ash accumulation rate prediction model, and calculate the current water-cooled wall ash accumulation rate data in combination with the ash melting point data corresponding to the ash content data;

[0009] Sootblowing demand prediction module: used to receive ash accumulation rate data and flue gas flow rate data, and calculate the remaining effective time data to reach the maximum allowable ash accumulation thickness based on the difference between the preset maximum allowable ash accumulation thickness data and the current actual ash accumulation thickness data, combined with the inhibitory effect of flue gas flow rate on ash accumulation rate;

[0010] Intelligent sootblowing decision module: used to match the corresponding sootblowing strategy according to the remaining effective time data, and then output a decision instruction set including sootblowing position, sootblowing execution timing and sootblowing intensity level.

[0011] Optionally, the water-wall operating condition synchronous acquisition module includes a temperature difference sensing unit, a coal quality data acquisition unit, and a flue gas flow rate monitoring unit; wherein:

[0012] Temperature difference sensing unit: This is installed in the corresponding area of ​​each panel of the boiler water wall and consists of a high-density distributed thermocouple sensor array. The sensor array is arranged in a group every 0.5 meters in the vertical direction, and temperature sensing nodes are set along each pipeline horizontally. The sensor collects the upstream and downstream inlet and outlet pipe wall temperature data in real time, and calculates the temperature difference of each panel pipe wall based on the difference algorithm. The collected temperature difference data is continuously output with a period of 1 second.

[0013] Coal quality data acquisition unit: This includes a gamma-ray coal quality online analysis device installed on the coal conveyor belt entering the furnace, which is used to continuously analyze the ash content of the coal sample before it enters the furnace. The analysis results are updated every 30 seconds and the coal ash content data at the corresponding moment is output;

[0014] Flue gas velocity monitoring unit: It is set in the flue gas mainstream channel above the water-cooled wall area. It collects the flue gas velocity at different heights during boiler operation in real time through a multi-point distributed Pitot tube array. The acquisition frequency is 10Hz, and it is synchronized and compared according to the sensor timestamp to output the flue gas velocity data aligned with the temperature difference sampling time.

[0015] Optionally, the temperature difference dynamic compensation module includes a humidity data acquisition unit, a compensation factor generation unit and a temperature difference correction calculation unit; wherein:

[0016] Humidity data acquisition unit: This unit is located in the external environment of the boiler water wall and uses a digital humidity sensor to collect real-time relative humidity at a frequency of 1 second. The collected results are automatically timestamped and aligned with the temperature difference data.

[0017] Compensation factor generation unit: used to calculate the humidity compensation factor based on the deviation between the real-time relative humidity value and the historical reference humidity value. The formula is: K = 1 + α (H - H0), where α is the humidity sensitivity coefficient calibrated based on the actual furnace conditions, H is the current relative humidity, and H0 is the standard humidity under normal operating conditions of the equipment;

[0018] Temperature difference correction calculation unit: used to convert the collected original pipe wall temperature difference data T d Multiply it by the compensation factor K to generate the corrected temperature difference value T after humidity compensation d ′, the compensation result is updated once per second and serves as the input data for the subsequent dust accumulation rate analysis module.

[0019] Optionally, the dust accumulation rate analysis module includes a temperature difference feature extraction unit, a melting point correlation calculation unit and a rate output unit; wherein:

[0020] Temperature difference feature extraction unit: used to receive the compensated temperature difference data output by the temperature difference dynamic compensation module, perform sliding statistical processing on the temperature difference sequence within a continuous time period based on a preset sampling window, and extract the temperature difference characteristic parameters corresponding to each water-cooled wall, including the temperature difference mean, temperature difference trend slope characteristics, and temperature difference fluctuation amplitude;

[0021] Melting point correlation calculation unit: used to query the ash melting point data of the corresponding coal type in the coal quality database based on the ash content data provided by the coal quality data acquisition unit, and construct the depositability weight factor based on the thermodynamic adaptation relationship between the current temperature difference characteristics and the coal ash melting point;

[0022] Rate output unit: used to input the temperature difference characteristic parameters and melting point weight factors into the pre-trained dust accumulation rate prediction model. After the model completes internal reasoning, it outputs the water-cooled wall dust accumulation rate data of the corresponding screen position under the current working conditions.

[0023] Optionally, the melting point correlation calculation unit includes:

[0024] Temperature difference normalization processing subunit: used to convert the water wall surface temperature difference mean parameter T output by the temperature difference feature extraction unit d ′ is converted to the ash melting point T m The thermal adaptability value of the same dimension is calculated by the maximum operating temperature T max Normalization processing to obtain the standard temperature difference value T n ;

[0025] Melting point matching evaluation subunit: used to calculate the relative adaptation coefficient R between the standard temperature difference value and the coal ash melting point. Its expression is:

[0026] Deposition factor generation subunit: used to generate the depositability weight factor W according to the relative adaptation coefficient R, which is specifically given by the following expression:

[0027] Optionally, the rate output unit includes:

[0028] Model input preprocessing subunit: used to receive the temperature difference mean, temperature difference fluctuation amplitude, temperature difference trend slope parameters output by the temperature difference feature extraction unit, and the deposition weight factor output by the melting point correlation calculation unit, and perform normalization according to the input vector format of the dust accumulation rate prediction model to generate a standardized input feature vector;

[0029] Dust accumulation rate model inference subunit: used to call the pre-trained dust accumulation rate prediction model, input the standardized input feature vector, and perform inference calculations based on the trained nonlinear regression function relationship within the model to obtain the current water-cooled wall dust accumulation rate prediction value;

[0030] Dust accumulation rate result output subunit: used to predict the dust accumulation rate V obtained by model reasoning rate Perform inverse normalization processing, restore to the physical unit mm / h, and output the water-cooled wall dust accumulation rate data of the corresponding screen position under the current working conditions.

[0031] Optionally, the soot blowing demand prediction module includes a soot thickness calculation subunit, a smoke velocity impact correction subunit, and a remaining time estimation subunit; wherein:

[0032] The dust accumulation thickness calculation subunit is used to receive the water-cooled wall dust accumulation rate data provided by the rate output unit, and calculate the current actual dust accumulation thickness value based on the time interval after the last soot blowing recorded in the system and the cumulative time of the current screen position. At the same time, it calls the maximum allowable dust accumulation thickness value of the corresponding screen position in the parameter management library to calculate the thickness difference ΔH between the two;

[0033] Smoke velocity effect correction subunit: It is used to receive the flue gas flow rate data provided by the water-cooled wall working condition synchronous acquisition module, and dynamically correct the ash deposition rate data based on the difference between the current flue gas flow rate and the historical reference flow rate value of the corresponding screen position, combined with the experimental calibration curve, to quantify the degree of inhibition of the flue gas flow rate on the ash deposition rate, and generate the corrected actual ash deposition rate V adj ;

[0034] Remaining time estimation subunit: used to calculate the time interval from the current moment to the time when the dust accumulation thickness reaches the set upper limit based on the dust accumulation thickness difference and the corrected actual dust accumulation rate, output the remaining effective time data under the corresponding screen position, and bind the corresponding timestamp structured output.

[0035] Optionally, the remaining time estimation subunit includes:

[0036] Thickness difference receiving subunit: used to receive the current thickness difference ΔH output by the dust accumulation thickness calculation subunit, and perform a numerical validity check on it to ensure that it is greater than zero and does not exceed the set maximum threshold range;

[0037] Rate matching calculation subunit: used to receive the corrected actual ash accumulation rate V output by the smoke velocity impact correction subunit adj , and calculate it together with the thickness difference ΔH to determine the time interval required to reach the maximum allowable dust accumulation thickness; the calculation formula is: Among them, t res Indicates the remaining effective time from the current moment to the maximum allowed dust accumulation limit;

[0038] Time interval output subunit: used to output the calculated remaining effective time t res Bind with the current timestamp, output structured data, mark the corresponding screen position number, remaining valid time and generation time.

[0039] Optionally, the intelligent sootblowing decision module includes a strategy matching unit, an instruction generation unit and an instruction output unit; wherein:

[0040] Strategy matching unit: used to receive the remaining effective time data of each screen position output by the remaining time estimation unit, and call the preset soot blowing strategy library to match the corresponding soot blowing strategy rules according to the preset time interval level of the remaining time. The soot blowing strategy rules include soot blowing trigger conditions, priority level, corresponding position number and recommended intensity parameters;

[0041] Instruction generation unit: used to generate decision instructions including sootblowing position, sootblowing execution time and sootblowing intensity level according to the matched sootblowing strategy rules, combined with the screen position number and the current system time, and encode the decision instructions into a data instruction set in a standard format;

[0042] Instruction output unit: used to push the decision instruction set to the boiler sootblowing control interface and simultaneously write it into the operation record database.

[0043] Optionally, the sootblowing strategy rules include:

[0044] Rule 1: When the remaining effective time is greater than or equal to 24 hours, the low-priority policy is matched and preventive soot blowing is performed on screen positions with an average temperature difference greater than 200°C, with the intensity level set to level 1.

[0045] Rule 2: When the remaining effective time is between 12 and 24 hours, match the medium priority strategy and select screen positions with a dust accumulation rate greater than 1.5 mm / h for planned soot blowing, setting the intensity level to level 2;

[0046] Rule 3: When the remaining effective time is between 4 and 12 hours, match the high-priority strategy and select screen positions with a remaining time less than 60% of the average remaining time for regional dust blowing, setting the intensity level to level 3;

[0047] Rule 4: When the remaining effective time is less than 4 hours, match the highest priority strategy and perform synchronous high-intensity dust blowing operations on all screen positions, setting the intensity level to level 4.

[0048] Beneficial effects of the present invention:

[0049] The present invention achieves accurate prediction of the water-cooled wall ash accumulation rate by synchronously collecting multi-source data such as tube wall temperature difference, ash melting point and flue gas flow rate, and quickly matches the graded sootblowing strategy in combination with the remaining effective time analysis, thereby realizing dynamic and efficient intelligent sootblowing decision-making under the fluctuating operating conditions of the coal-fired unit, and improving the stability and economy of boiler operation. With the help of this solution, excessive or insufficient operation in the timed sootblowing mode can be avoided, reducing unnecessary energy consumption and the risk of shutdown.

[0050] Based on real-time monitoring and big data modeling, the present invention deeply couples coal characteristics, thermal conditions and boiler demand. Through sootblowing demand prediction and strategy optimization, it effectively coordinates the load and maintenance frequency of each water-cooled wall screen, reduces local temperature imbalance and tube wall damage caused by soot accumulation, and ensures long-term safe and efficient operation of the unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0052] Figure 1 Schematic diagram of an intelligent operation auxiliary decision-making system for thermal power units according to an embodiment of the present invention;

[0053] Figure 2 Schematic diagram of a dust accumulation rate analysis module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. It is also noted that, to provide a more detailed description, the following embodiments are best and preferred embodiments, and those skilled in the art may employ alternative methods for implementing certain known technologies. Furthermore, the accompanying drawings are intended only to provide a more detailed description of the embodiments and are not intended to limit the present invention.

[0055] It should be noted that references in the specification to "one embodiment," "an embodiment," "an exemplary embodiment," "some embodiments," etc. indicate that the described embodiments may include specific features, structures, or characteristics, but not every embodiment necessarily includes such specific features, structures, or characteristics. In addition, when specific features, structures, or characteristics are described in conjunction with an embodiment, it is within the knowledge of persons skilled in the relevant art to implement such features, structures, or characteristics in conjunction with other embodiments (whether or not explicitly described).

[0056] In general, terms can be understood, at least in part, from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending at least in part on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0057] like Figure 1-Figure 2 As shown, an intelligent operation auxiliary decision-making system for thermal power units based on big data analysis includes a water-wall operating condition synchronous acquisition module, a temperature difference dynamic compensation module, a soot accumulation rate analysis module, a soot blowing demand prediction module, and an intelligent soot blowing decision-making module; wherein:

[0058] Water-wall operating condition synchronous acquisition module: It is used to collect the temperature difference data of each panel tube wall in real time through the distributed temperature sensor array arranged on each panel of the boiler water-wall, and synchronously obtain the ash content data of the coal entering the furnace and the flue gas flow rate data at the corresponding time;

[0059] Temperature difference dynamic compensation module: used to receive tube wall temperature difference data, ash content data and flue gas flow rate data, and perform humidity compensation calculation on the tube wall temperature difference data based on the real-time humidity data obtained by the ambient humidity sensor, and output the compensated temperature difference data;

[0060] Ash accumulation rate analysis module: used to input the compensated temperature difference data into the ash accumulation rate prediction model, and calculate the current water-cooled wall ash accumulation rate data in combination with the ash melting point data corresponding to the ash content data;

[0061] Sootblowing demand prediction module: used to receive ash accumulation rate data and flue gas flow rate data, and calculate the remaining effective time data to reach the maximum allowable ash accumulation thickness based on the difference between the preset maximum allowable ash accumulation thickness data and the current actual ash accumulation thickness data, combined with the inhibitory effect of flue gas flow rate on ash accumulation rate;

[0062] Intelligent sootblowing decision module: used to match the corresponding sootblowing strategy according to the remaining effective time data, and then output a decision instruction set including sootblowing position, sootblowing execution timing and sootblowing intensity level.

[0063] The water-cooled wall working condition synchronous acquisition module includes a temperature difference sensing unit, a coal quality data acquisition unit, and a flue gas flow rate monitoring unit; among which:

[0064] Temperature difference sensing unit: This is installed in the corresponding area of ​​each panel of the boiler water wall and consists of a high-density distributed thermocouple sensor array. The sensor array is arranged in a group every 0.5 meters in the vertical direction, and temperature sensing nodes are set along each pipeline horizontally. The sensor collects the upstream and downstream inlet and outlet pipe wall temperature data in real time, and calculates the temperature difference of each panel pipe wall based on the difference algorithm. The collected temperature difference data is continuously output with a period of 1 second.

[0065] Coal quality data acquisition unit: This includes a gamma-ray coal quality online analysis device installed on the coal conveyor belt entering the furnace, which is used to continuously analyze the ash content of the coal sample before it enters the furnace. The analysis results are updated every 30 seconds and the coal ash content data at the corresponding moment is output;

[0066] Flue gas flow rate monitoring unit: It is set in the mainstream flue gas channel above the water-cooled wall area. The flue gas flow rate at different heights during boiler operation is collected in real time through a multi-point distributed Pitot tube array. The collection frequency is 10Hz, and synchronous comparison is performed according to the sensor timestamp to output flue gas flow rate data aligned with the temperature difference sampling time. By setting the temperature difference sensing unit, coal quality data acquisition unit and flue gas flow rate monitoring unit in the key thermal channel, coal inlet path and main flue gas flow channel respectively, not only the high-frequency coordinated collection of water-cooled wall temperature difference data and key boiler operation parameters is achieved, but also the consistency of the data in space and time is ensured, thereby enhancing the accuracy of subsequent dynamic compensation and decision-making calculations.

[0067] The temperature difference dynamic compensation module includes a humidity data acquisition unit, a compensation factor generation unit and a temperature difference correction calculation unit; wherein:

[0068] Humidity data acquisition unit: This unit is located in the external environment of the boiler water wall and uses a digital humidity sensor to collect real-time relative humidity at a frequency of 1 second. The collected results are automatically timestamped and aligned with the temperature difference data.

[0069] Compensation factor generation unit: used to calculate the humidity compensation factor based on the deviation between the real-time relative humidity value and the historical reference humidity value. The formula is: K = 1 + α (H - H0), where α is the humidity sensitivity coefficient calibrated based on the actual furnace conditions, H is the current relative humidity, and H0 is the standard humidity under normal operating conditions of the equipment;

[0070] The specific calculation formula of humidity sensitivity coefficient α is:

[0071] Where: T d,ref is the water wall temperature difference data corresponding to the standard reference working condition (relative humidity is H0); Td,wet The current high humidity condition (relative humidity is H wet ) water wall temperature difference data measured under H wet is the measured relative humidity of the current high humidity condition;

[0072] Temperature difference correction calculation unit: used to convert the collected original pipe wall temperature difference data T d Multiply it by the compensation factor K to generate the corrected temperature difference value T after humidity compensation d ′, the calculation formula is: T d ′=T d ×K, the compensation result is updated once per second and serves as the input data for the subsequent dust accumulation rate analysis module; by converting the actual humidity deviation value into a dynamic compensation factor and performing a multiplicative correction operation in combination with the original temperature difference data, it can effectively eliminate the interference of ambient humidity fluctuations on temperature difference measurement, improve the reliability of temperature difference data under high humidity or low humidity conditions, and provide more accurate thermal basic data for subsequent dust accumulation rate analysis.

[0073] The ash accumulation rate analysis module includes a temperature difference feature extraction unit, a melting point correlation calculation unit, and a rate output unit; wherein:

[0074] Temperature difference feature extraction unit: used to receive the compensated temperature difference data output by the temperature difference dynamic compensation module, perform sliding statistical processing on the temperature difference sequence within a continuous time period based on a preset sampling window, and extract the temperature difference characteristic parameters corresponding to each water-cooled wall, including the temperature difference mean, temperature difference trend slope characteristics, and temperature difference fluctuation amplitude, as a representation of the thermal changes in ash deposition;

[0075] Melting point correlation calculation unit: used to query the ash melting point data of the corresponding coal type in the coal quality database based on the ash content data provided by the coal quality data acquisition unit, and construct a depositability weight factor based on the thermodynamic adaptation relationship between the current temperature difference characteristics and the ash melting point to indicate whether the coal ash is easy to adhere to and fuse on the current water-cooled wall surface;

[0076] Table 1 Coal quality database structure

[0077]

[0078] When the system is running, the coal quality data acquisition unit will output the measured ash content of the current coal sample. The system will match the value with the ash content range field, select the corresponding coal type number and coal ash melting point, and pass it to the melting point matching evaluation subunit to further participate in the construction of the deposition weight factor.

[0079] Rate output unit: used to input the temperature difference characteristic parameters and melting point weight factors into the pre-trained ash accumulation rate prediction model. After the model completes internal reasoning, it outputs the water-cooled wall ash accumulation rate data of the corresponding screen position under the current working conditions, and structured encodes the results for the soot blowing demand prediction module to call; the above unit can dynamically judge the actual ash formation trend of coal ash on the water-cooled wall surface by associating the temperature difference change characteristics with the coal ash melting point data, so that the ash accumulation rate prediction model has both thermal response capabilities and coal combustion characteristics adaptability, thereby improving the prediction accuracy under complex coupling conditions of different coal types.

[0080] Melting point correlation calculation unit includes:

[0081] Temperature difference normalization processing subunit: used to convert the water wall surface temperature difference mean parameter T output by the temperature difference feature extraction unit d ′ is converted to the ash melting point T m The thermal adaptability value of the same dimension is calculated by the maximum operating temperature T max Normalization processing to obtain the standard temperature difference value T n , whose expression is: Among them, T max The upper safety limit wall temperature set for boiler operation;

[0082] Melting point matching evaluation subunit: used to calculate the relative adaptation coefficient R between the standard temperature difference value and the coal ash melting point. Its expression is: The coefficient R reflects whether the current wall temperature field is in the critical range of coal ash adhesion. The closer the value is to 1, the easier it is to form deposition.

[0083] Deposition factor generation subunit: used to generate the depositability weight factor W according to the relative adaptation coefficient R, which is specifically given by the following expression: When the adaptation coefficient R exceeds the set threshold of 0.6, it is considered that the current thermal conditions have the possibility of forming coal ash deposition, and the weight factor W is R; otherwise, it is judged to be a non-deposition state, and the output weight is 0; the above subunit matches and evaluates the normalized temperature difference with the coal ash melting point, and sets the adaptation coefficient threshold to construct the deposition factor, which can realize the quantitative judgment of the thermal adhesion trend of the coal ash, provide key weight parameters based on the fusion of thermodynamics and physical properties for the prediction of ash deposition rate, and help improve the prediction stability of the model under the conditions of multi-coal adaptation.

[0084] The rate output unit includes:

[0085] Model input preprocessing subunit: used to receive the temperature difference mean, temperature difference fluctuation amplitude, temperature difference trend slope parameters output by the temperature difference feature extraction unit, and the deposition weight factor output by the melting point correlation calculation unit, and perform normalization according to the input vector format of the dust accumulation rate prediction model to generate a standardized input feature vector;

[0086] Dust accumulation rate model inference subunit: used to call the pre-trained dust accumulation rate prediction model, input the standardized input feature vector, and perform inference calculations based on the trained nonlinear regression function relationship within the model to obtain the current water-cooled wall dust accumulation rate prediction value;

[0087] The specific reasoning process is as follows:

[0088] Assume the normalized input feature vector is: X = [T avg , T amp , T slope , W], where T avg is the normalized temperature difference mean characteristic; T amp is the normalized temperature difference fluctuation amplitude characteristic; T slope is the normalized temperature difference trend slope characteristic; W is the sedimentation weight factor;

[0089] The weight matrix obtained by pre-training the dust accumulation rate model is M, and the bias vector is B. The current water-cooled wall dust accumulation rate prediction value V rate The calculation expression is: V rate =f(MX T +B), where f(·) is the activation function, using the ReLU function, and is defined as: f(x) = max(0, x);

[0090] Dust accumulation rate result output subunit: used to predict the dust accumulation rate V obtained by model reasoning rate The inverse normalization process is performed to restore the data to the physical unit mm / h, and the water-cooled wall dust accumulation rate data of the corresponding screen position under the current working conditions is output for the soot blowing demand prediction module to call. The above subunits realize the fine integration of various characteristics and physical mechanisms through the standardized preprocessing of input features and the nonlinear reasoning calculation of the dust accumulation rate model, thereby improving the real-time and accuracy of the dust accumulation rate prediction.

[0091] The soot blowing demand prediction module includes a soot thickness calculation subunit, a smoke velocity impact correction subunit, and a remaining time estimation subunit; wherein:

[0092] The dust accumulation thickness calculation subunit is used to receive the water-cooled wall dust accumulation rate data provided by the rate output unit, and calculate the current actual dust accumulation thickness value based on the time interval after the last soot blowing recorded in the system and the cumulative time of the current screen position. At the same time, it calls the maximum allowable dust accumulation thickness value of the corresponding screen position in the parameter management library to calculate the thickness difference ΔH between the two;

[0093] Current actual dust accumulation thickness H cur The calculation formula is: cur =V rate Δt, where V rateis the dust accumulation rate before the current screen position correction (unit: mm / h), Δt is the running time from the last soot blowing (unit: h); the calculation formula of the thickness difference ΔH is: ΔH=H max -H cur , where H max The maximum allowable dust accumulation thickness preset in the system (unit: mm);

[0094] Smoke velocity effect correction subunit: It is used to receive the flue gas flow rate data provided by the water-cooled wall working condition synchronous acquisition module, and dynamically correct the ash deposition rate data based on the difference between the current flue gas flow rate and the historical reference flow rate value of the corresponding screen position, combined with the experimental calibration curve, to quantify the degree of inhibition of the flue gas flow rate on the ash deposition rate, and generate the corrected actual ash deposition rate V adj ;

[0095] The specific calculation formula is: Among them, V flue is the current screen position flue gas velocity (unit: m / s), V ref is the historical reference flow rate of the corresponding screen position (unit: m / s), β is the flow rate sensitivity coefficient determined based on the measured data;

[0096] The velocity sensitivity coefficient β reflects the inhibitory effect of flue gas velocity changes on ash deposition rate, and its calculation formula is as follows: Among them, V rate,ref Represents the reference flow rate V ref The dust accumulation rate measured under the conditions (unit: mm / h); V rate,high Represents the high flow rate V high The dust accumulation rate measured under the conditions (unit: mm / h); V ref is the reference flow rate of the flue gas at the screen position under historical working conditions (unit: m / s); V high is the actual measured flue gas flow rate higher than the reference flow rate (unit: m / s);

[0097] Remaining time estimation subunit: It is used to calculate the time interval from the current moment to the time when the ash thickness reaches the set upper limit based on the difference in ash thickness and the corrected actual ash accumulation rate, output the remaining effective time data under the corresponding screen position, and bind the corresponding timestamp structured output for the intelligent soot blowing decision module to call; the above unit introduces the ash thickness calculation and smoke speed correction mechanism with time as the variable, which can dynamically adapt to the fluctuations of the flue gas flow field and ash deposition characteristics, improve the calculation accuracy of the remaining ash tolerance time, and provide high reliability support for the intelligent soot blowing timing determination.

[0098] The remaining time estimation subunit includes:

[0099] Thickness difference receiving subunit: used to receive the current thickness difference ΔH output by the dust accumulation thickness calculation subunit, and perform a numerical validity check on it to ensure that it is greater than zero and does not exceed the set maximum threshold range;

[0100] Rate matching calculation subunit: used to receive the corrected actual ash accumulation rate V output by the smoke velocity impact correction subunit adj , and calculate it together with the thickness difference ΔH to determine the time interval required to reach the maximum allowable dust accumulation thickness; the calculation formula is: Among them, t res Indicates the remaining effective time from the current moment to reaching the upper limit of allowed dust accumulation (unit: h);

[0101] Time interval output subunit: used to output the calculated remaining effective time t res It is bound to the current timestamp, outputs structured data, and marks the corresponding screen position number, remaining effective time and generation time for the intelligent sootblowing decision module to retrieve and judge; through the real-time ratio calculation of thickness difference and correction rate, the system can accurately calculate the time margin of each screen water-cooled wall from the critical soot accumulation state, forming a dynamically updated decision-making reference, which is helpful to plan the sootblowing rhythm in advance and ensure the thermal efficiency of boiler operation and the safety redundant space of the tube wall.

[0102] The intelligent sootblowing decision module includes a strategy matching unit, an instruction generation unit, and an instruction output unit; wherein:

[0103] Strategy matching unit: used to receive the remaining effective time data of each screen position output by the remaining time estimation unit, and call the preset soot blowing strategy library to match the corresponding soot blowing strategy rules according to the preset time interval level of the remaining time. The soot blowing strategy rules include soot blowing trigger conditions, priority level, corresponding position number and recommended intensity parameters;

[0104] Instruction generation unit: used to generate decision instructions including sootblowing position, sootblowing execution time and sootblowing intensity level according to the matched sootblowing strategy rules, combined with the screen position number and the current system time, and encode the decision instructions into a data instruction set in a standard format;

[0105] Instruction output unit: used to push the decision instruction set to the boiler sootblowing control interface and simultaneously write it into the operation record database to facilitate operation backtracking and scheduling coordination, and display the decision execution plan on the scheduling system interface; the above unit establishes a mapping relationship between the remaining time and the strategy library rules to achieve dynamic strategy matching based on segmentation, timeliness and intensity, and form a structured decision instruction set to improve the timeliness, accuracy and safety of sootblowing operations, providing highly responsive control support for intelligent optimization of boiler operations.

[0106] Sootblowing strategy rules include:

[0107] Rule 1: When the remaining effective time is greater than or equal to 24 hours, the low-priority policy is matched and preventive soot blowing is performed on screen positions with an average temperature difference greater than 200°C, with the intensity level set to level 1.

[0108] Rule 2: When the remaining effective time is between 12 and 24 hours, match the medium priority strategy and select screen positions with a dust accumulation rate greater than 1.5 mm / h for planned soot blowing, setting the intensity level to level 2;

[0109] Rule 3: When the remaining effective time is between 4 and 12 hours, match the high-priority strategy and select screen positions with a remaining time less than 60% of the average remaining time for regional dust blowing, setting the intensity level to level 3;

[0110] Rule 4: When the remaining effective time is less than 4 hours, match the highest priority strategy and perform synchronous high-intensity soot blowing operations on all screen positions, setting the intensity level to level 4. The above rules set clear temperature difference, soot accumulation rate and relative remaining time thresholds, enabling the system to automatically execute logical judgment conditions when the strategy is matched, thereby improving the accuracy, timeliness and stability of soot blowing scheduling.

[0111] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0112] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. An intelligent operation auxiliary decision-making system for thermal power units based on big data analysis, characterized in that: It includes a water-cooled wall operating condition synchronous acquisition module, a temperature difference dynamic compensation module, a soot accumulation rate analysis module, a soot blowing demand prediction module, and an intelligent soot blowing decision module; among which: Water-wall operating condition synchronous acquisition module: It is used to collect the temperature difference data of each panel tube wall in real time through the distributed temperature sensor array arranged on each panel of the boiler water-wall, and synchronously obtain the ash content data of the coal entering the furnace and the flue gas flow rate data at the corresponding time; Temperature difference dynamic compensation module: used to receive tube wall temperature difference data, ash content data and flue gas flow rate data, and perform humidity compensation calculation on the tube wall temperature difference data based on the real-time humidity data obtained by the ambient humidity sensor, and output the compensated temperature difference data; Ash accumulation rate analysis module: used to input the compensated temperature difference data into the ash accumulation rate prediction model, and calculate the current water-cooled wall ash accumulation rate data in combination with the ash melting point data corresponding to the ash content data; Sootblowing demand prediction module: used to receive ash accumulation rate data and flue gas flow rate data, and calculate the remaining effective time data to reach the maximum allowable ash accumulation thickness based on the difference between the preset maximum allowable ash accumulation thickness data and the current actual ash accumulation thickness data, combined with the inhibitory effect of flue gas flow rate on ash accumulation rate; Intelligent sootblowing decision module: used to match the corresponding sootblowing strategy according to the remaining effective time data, and then output a decision instruction set including sootblowing position, sootblowing execution timing and sootblowing intensity level.

2. The intelligent operation auxiliary decision-making system for thermal power units based on big data analysis according to claim 1 is characterized in that: The water-cooled wall working condition synchronous acquisition module includes a temperature difference sensing unit, a coal quality data acquisition unit and a flue gas flow rate monitoring unit; wherein: Temperature difference sensing unit: This is installed in the corresponding area of ​​each panel of the boiler water wall and consists of a high-density distributed thermocouple sensor array. The sensor array is arranged in a group every 0.5 meters in the vertical direction, and temperature sensing nodes are set along each pipeline horizontally. The sensor collects the upstream and downstream inlet and outlet pipe wall temperature data in real time, and calculates the temperature difference of each panel pipe wall based on the difference algorithm. The collected temperature difference data is continuously output with a period of 1 second. Coal quality data acquisition unit: This includes a gamma-ray coal quality online analysis device installed on the coal conveyor belt entering the furnace, which is used to continuously analyze the ash content of the coal sample before it enters the furnace. The analysis results are updated every 30 seconds and the coal ash content data at the corresponding moment is output; Flue gas velocity monitoring unit: It is set in the flue gas mainstream channel above the water-cooled wall area. It collects the flue gas velocity at different heights during boiler operation in real time through a multi-point distributed Pitot tube array. The acquisition frequency is 10Hz, and it is synchronized and compared according to the sensor timestamp to output the flue gas velocity data aligned with the temperature difference sampling time.

3. The intelligent operation auxiliary decision-making system for thermal power units based on big data analysis according to claim 1 is characterized in that: The temperature difference dynamic compensation module includes a humidity data acquisition unit, a compensation factor generation unit and a temperature difference correction calculation unit; wherein: Humidity data acquisition unit: This unit is located in the external environment of the boiler water wall and uses a digital humidity sensor to collect real-time relative humidity at a frequency of 1 second. The collected results are automatically timestamped and aligned with the temperature difference data. Compensation factor generation unit: used to calculate the humidity compensation factor based on the deviation between the real-time relative humidity value and the historical reference humidity value. The formula is: K = 1 + α (H - H0), where α is the humidity sensitivity coefficient calibrated based on the actual furnace conditions, H is the current relative humidity, and H0 is the standard humidity under normal operating conditions of the equipment; Temperature difference correction calculation unit: used to convert the collected original pipe wall temperature difference data T d Multiply it by the compensation factor K to generate the corrected temperature difference value T after humidity compensation d ′, the compensation result is updated once per second and serves as the input data for the subsequent dust accumulation rate analysis module.

4. The intelligent operation auxiliary decision-making system for thermal power units based on big data analysis according to claim 2 is characterized in that: The dust accumulation rate analysis module includes a temperature difference feature extraction unit, a melting point correlation calculation unit and a rate output unit; wherein: Temperature difference feature extraction unit: used to receive the compensated temperature difference data output by the temperature difference dynamic compensation module, perform sliding statistical processing on the temperature difference sequence within a continuous time period based on a preset sampling window, and extract the temperature difference characteristic parameters corresponding to each water-cooled wall, including the temperature difference mean, temperature difference trend slope characteristics, and temperature difference fluctuation amplitude; Melting point correlation calculation unit: used to query the ash melting point data of the corresponding coal type in the coal quality database based on the ash content data provided by the coal quality data acquisition unit, and construct the depositability weight factor based on the thermodynamic adaptation relationship between the current temperature difference characteristics and the coal ash melting point; Rate output unit: used to input the temperature difference characteristic parameters and melting point weight factors into the pre-trained dust accumulation rate prediction model. After the model completes internal reasoning, it outputs the water-cooled wall dust accumulation rate data of the corresponding screen position under the current working conditions.

5. The intelligent operation auxiliary decision-making system for thermal power units based on big data analysis according to claim 4 is characterized in that: The melting point correlation calculation unit includes: Temperature difference normalization processing subunit: used to convert the water wall surface temperature difference mean parameter T output by the temperature difference feature extraction unit d ′ is converted to the ash melting point T m The thermal adaptability value of the same dimension is calculated by the maximum operating temperature T max Normalization processing to obtain the standard temperature difference value T n ; Melting point matching evaluation subunit: used to calculate the relative adaptation coefficient R between the standard temperature difference value and the coal ash melting point. Its expression is: Deposition factor generation subunit: used to generate the depositability weight factor W according to the relative adaptation coefficient R, which is specifically given by the following expression:

6. The intelligent operation auxiliary decision-making system for thermal power units based on big data analysis according to claim 5 is characterized in that: The rate output unit includes: Model input preprocessing subunit: used to receive the temperature difference mean, temperature difference fluctuation amplitude, temperature difference trend slope parameters output by the temperature difference feature extraction unit, and the deposition weight factor output by the melting point correlation calculation unit, and perform normalization according to the input vector format of the dust accumulation rate prediction model to generate a standardized input feature vector; Dust accumulation rate model inference subunit: used to call the pre-trained dust accumulation rate prediction model, input the standardized input feature vector, and perform inference calculations based on the trained nonlinear regression function relationship within the model to obtain the current water-cooled wall dust accumulation rate prediction value; Dust accumulation rate result output subunit: used to predict the dust accumulation rate V obtained by model reasoning rate Perform inverse normalization processing, restore to the physical unit mm / h, and output the water-cooled wall dust accumulation rate data of the corresponding screen position under the current working conditions.

7. The intelligent operation auxiliary decision-making system for thermal power units based on big data analysis according to claim 1 is characterized in that: The soot blowing demand prediction module includes a soot thickness calculation subunit, a smoke velocity impact correction subunit and a remaining time estimation subunit; wherein: The dust accumulation thickness calculation subunit is used to receive the water-cooled wall dust accumulation rate data provided by the rate output unit, and calculate the current actual dust accumulation thickness value based on the time interval after the last soot blowing recorded in the system and the cumulative time of the current screen position. At the same time, it calls the maximum allowable dust accumulation thickness value of the corresponding screen position in the parameter management library to calculate the thickness difference ΔH between the two; Smoke velocity effect correction subunit: It is used to receive the flue gas flow rate data provided by the water-cooled wall working condition synchronous acquisition module, and dynamically correct the ash deposition rate data based on the difference between the current flue gas flow rate and the historical reference flow rate value of the corresponding screen position, combined with the experimental calibration curve, to quantify the degree of inhibition of the flue gas flow rate on the ash deposition rate, and generate the corrected actual ash deposition rate V adj ; Remaining time estimation subunit: used to calculate the time interval from the current moment to the time when the dust accumulation thickness reaches the set upper limit based on the dust accumulation thickness difference and the corrected actual dust accumulation rate, output the remaining effective time data under the corresponding screen position, and bind the corresponding timestamp structured output.

8. The intelligent operation auxiliary decision-making system for thermal power units based on big data analysis according to claim 7 is characterized in that: The remaining time estimation subunit includes: Thickness difference receiving subunit: used to receive the current thickness difference ΔH output by the dust accumulation thickness calculation subunit, and perform a numerical validity check on it to ensure that it is greater than zero and does not exceed the set maximum threshold range; Rate matching calculation subunit: used to receive the corrected actual ash accumulation rate V output by the smoke velocity impact correction subunit adj , and calculate it together with the thickness difference ΔH to determine the time interval required to reach the maximum allowable dust accumulation thickness; the calculation formula is: Among them, t res Indicates the remaining effective time from the current moment to the maximum allowed dust accumulation limit; Time interval output subunit: used to output the calculated remaining effective time t res Bind with the current timestamp, output structured data, mark the corresponding screen position number, remaining valid time and generation time.

9. The intelligent operation auxiliary decision-making system for thermal power units based on big data analysis according to claim 1 is characterized in that: The intelligent sootblowing decision module includes a strategy matching unit, an instruction generation unit and an instruction output unit; wherein: Strategy matching unit: used to receive the remaining effective time data of each screen position output by the remaining time estimation unit, and call the preset soot blowing strategy library to match the corresponding soot blowing strategy rules according to the preset time interval level of the remaining time. The soot blowing strategy rules include soot blowing trigger conditions, priority level, corresponding position number and recommended intensity parameters; Instruction generation unit: used to generate decision instructions including sootblowing position, sootblowing execution time and sootblowing intensity level according to the matched sootblowing strategy rules, combined with the screen position number and the current system time, and encode the decision instructions into a data instruction set in a standard format; Instruction output unit: used to push the decision instruction set to the boiler sootblowing control interface and simultaneously write it into the operation record database.

10. The intelligent operation auxiliary decision-making system for thermal power units based on big data analysis according to claim 9 is characterized in that: The sootblowing strategy rules include: Rule 1: When the remaining effective time is greater than or equal to 24 hours, the low-priority policy is matched and preventive soot blowing is performed on screen positions with an average temperature difference greater than 200°C, with the intensity level set to level 1. Rule 2: When the remaining effective time is between 12 and 24 hours, match the medium priority strategy and select screen positions with a dust accumulation rate greater than 1.5 mm / h for planned soot blowing, setting the intensity level to level 2; Rule 3: When the remaining effective time is between 4 and 12 hours, match the high-priority strategy and select screen positions with a remaining time less than 60% of the average remaining time for regional dust blowing, setting the intensity level to level 3; Rule 4: When the remaining effective time is less than 4 hours, match the highest priority strategy and perform synchronous high-intensity dust blowing operations on all screen positions, setting the intensity level to level 4.

Images