Real-time monitoring system and method for fly ash processing line based on multi-sensor fusion

Abstract

The application discloses a fly ash treatment production line real-time monitoring system and method based on multi-sensor fusion, relates to the technical field of fly ash treatment production line monitoring, and comprises the following steps: based on a fly ash treatment process, stages of a fly ash treatment production line are divided, after the division is completed, input parameters of each stage are obtained, the input parameters are output by an upstream stage, and output parameters of the input parameters are transmitted to a downstream stage; data information of each stage of the fly ash treatment production line is collected through a sensor, and data of each stage is respectively composed into a data set; by comprehensively considering weights, real-time values, target values and upper limits of running deviations of monitored physical parameters in the stage, a safety risk assessment value of the stage under a current running state is calculated; a risk transmission coefficient is calculated, a dynamic adjustment model is established by analyzing a relationship between the calculated risk transmission coefficient and a product failure rate, and fly ash treatment production line real-time monitoring is realized.

Description

Technical Field

[0001] This invention relates to the field of fly ash processing production line monitoring technology, specifically a real-time monitoring system and method for fly ash processing production lines based on multi-sensor fusion. Background Technology

[0002] With increasingly stringent environmental standards and ever-increasing requirements for the safe operation of fly ash treatment production lines, fly ash treatment technology is becoming increasingly important in the field of waste treatment. As a key to ensuring the stability and compliance of the fly ash treatment process, the real-time monitoring effect of fly ash treatment production lines is related to environmental safety.

[0003] However, traditional fly ash treatment line monitoring methods often face the following problems when dealing with complex treatment processes, dynamic changes in multiple parameters, and risk assessments: First, the monitoring data collection is incomplete and has low accuracy. Fly ash treatment involves multiple physical and chemical processes, generating a large amount of complex data. Traditional monitoring relies on a small number of sensors with limited accuracy, making it difficult to comprehensively acquire key parameters. Second, the data analysis and risk assessment capabilities are weak. Traditional monitoring often involves simple data recording and manual analysis, making it difficult to uncover potential risks and correlations behind the data. As the complexity of the treatment process increases, the interactions between parameters become more complex, and traditional methods cannot detect anomalies in a timely manner from dynamic changes in multiple parameters. In addition, there is a lack of real-time dynamic adjustment mechanisms. Traditional monitoring systems cannot adjust production line operating parameters in a timely manner based on real-time monitoring data. To address these issues, while existing fly ash treatment line monitoring systems have achieved automation to some extent, they still have significant shortcomings, especially when facing complex processes and dynamic changes in multiple parameters. They lack intelligent real-time monitoring, risk warning, and dynamic optimization capabilities, failing to meet the growing environmental and production demands. Summary of the Invention

[0004] The purpose of this invention is to provide a real-time monitoring system and method for fly ash processing production lines based on multi-sensor fusion, so as to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a real-time monitoring method for fly ash processing production lines based on multi-sensor fusion, the method comprising the following steps:

[0006] Based on the fly ash treatment process, the fly ash treatment production line is divided into stages. After the division is completed, the input parameters of each stage that receive the output of the upstream stage and the output parameters that are transmitted to the downstream stage are obtained.

[0007] Data from each stage of the fly ash processing production line is collected by sensors, and the data from each stage is compiled into a dataset.

[0008] By comprehensively considering the weights, real-time values, target values, and upper limits of operational deviations of the physical parameters monitored during the phase, the safety risk assessment value of this phase under the current operational state is calculated.

[0009] Based on the real-time parameters of each stage of the fly ash treatment production line, upstream and downstream parameter combinations are extracted. Highly correlated parameter combinations are selected through gray relational degree calculation. The risk transmission coefficient is obtained by calculating the partial derivatives in combination with the process mechanism and introducing time decay correction. The relationship between the coefficient and the product defect rate is analyzed to establish a dynamic adjustment model.

[0010] Based on the fly ash treatment process, the fly ash treatment production line is divided into stages. After the division, the input parameters of each stage that receive the output from the upstream stage and the output parameters that are passed to the downstream stage are obtained. The specific steps include:

[0011] The fly ash treatment production line is divided into n stages according to the treatment process, denoted as N={N1,N2,...,Nn}, where N1,N2,...,Nn represent the corresponding 1st, 2nd,...,nth fly ash treatment production line stages, and n represents the total number of stages into which the fly ash treatment production line is divided.

[0012] After dividing the fly ash treatment production line into stages, each stage is marked with i. The input and output parameters of each stage are obtained. The input parameters of each stage are inherited from the output of the upstream stage, and the output parameters are passed to the downstream stage. Here, stage i represents the i-th fly ash treatment stage.

[0013] Data from each stage of the fly ash treatment production line is collected using sensors, and the data from each stage is compiled into a dataset. The specific steps include:

[0014] Data information from each stage of the fly ash treatment production line is acquired through sensors. The acquired data information from each stage of the fly ash treatment production line is cleaned and preprocessed. When a data value is missing at a certain moment, a time series-based interpolation method is used to fill it in based on the data values ​​of the preceding and following moments. The data format is standardized and all data is stored in a unified database table structure.

[0015] The Min-Max normalization algorithm is used to process the data information of each stage of the fly ash processing production line and map the data to the [0,1] interval.

[0016] The cleaned data from each stage are grouped into datasets, denoted as {x1,x2,...,xn}. The dataset x1 from the first stage contains m1 parameters and is denoted as {x(1,1),x(1,2),...,x(1,m1)}. The dataset x2 from the second stage contains m2 parameters and is denoted as {x(2,1),x(2,2),...,x(2,m2)}, and so on. The dataset xn from the nth stage contains mn parameters and is denoted as {x(n,1),x(n,2),...,x(n,m)}. n Let x(1,1), x(1,2), ..., x(1,m1) represent the 1st, 2nd, ..., m1st parameters of the 1st stage, x(2,1), x(2,2), ..., x(2,m2) represent the 1st, 2nd, ..., m2nd parameters of the 2nd stage, and so on, x(n,1), x(n,2), ..., x(n,m) n ) represent the nth stage, 1, 2, ..., m respectively. n There are several parameters, m1, m2, ..., m n These represent the total number of parameters in the 1st, 2nd, ..., nth stages, respectively.

[0017] By comprehensively considering the weights, real-time values, target values, and upper limits of operational deviations of the physical parameters monitored during the phase, the safety risk assessment value of this phase under the current operational state is calculated. The specific steps include:

[0018] A risk assessment function is defined for each stage, as shown below:

[0019] ;

[0020] Among them, R i This represents the safety risk assessment value for the i-th stage, reflecting the comprehensive degree of anomaly of all parameters in that stage. i w represents the total number of parameters monitored in the i-th stage. i,k Let x(i,k) represent the weight of the k-th parameter in the i-th stage. This weight is determined through training with historical data. Let x(i,k) represent the real-time measured value of the k-th parameter in the i-th stage. Let T(i,k) represent the target value of the k-th parameter in the i-th stage, which is the ideal state required by the process design. Let H(i,k) represent the upper limit of the operating deviation of the k-th parameter in the i-th stage, which is the maximum tolerable range of the actual value deviating from the target value.

[0021] Based on real-time parameters at each stage of the fly ash treatment production line, upstream and downstream parameter combinations are extracted. Highly correlated parameter combinations are selected through grey relational analysis. Partial derivatives are calculated using the process mechanism, and a risk transmission coefficient is obtained by introducing time decay correction. The relationship between this coefficient and the product defect rate is analyzed to establish a dynamic adjustment model. Specific steps include:

[0022] Based on real-time parameter data of each stage of the fly ash treatment production line, parameter combinations between upstream and downstream stages are extracted. According to the grey relational degree, the grey relational coefficient γ(x(i,k),x(j,l)) of the parameter combinations between upstream and downstream stages is calculated. By setting a correlation strength threshold, parameter combinations whose grey relational coefficients exceed the set correlation strength threshold are filtered out. Here, γ(x(i,k),x(j,l)) represents the grey relational coefficient between parameter x(i,k) of stage Ni and parameter x(j,l) of stage Nj, which measures the statistical correlation strength between parameter x(i,k) of stage Ni and parameter x(j,l) of stage Nj.

[0023] The correlation coefficient is calculated using grey relational analysis. The formula for grey relational analysis is as follows:

[0024] ;

[0025] Where γ(x(i,k),x(j,l)) represents the statistical correlation strength between the real-time values ​​of the k-th parameter of stage Ni and the real-time values ​​of the l-th parameter of stage Nj, and ρ represents the resolution coefficient.

[0026] Based on the parameter combinations whose gray correlation coefficients exceed the set correlation strength threshold, the partial derivatives of downstream parameters with respect to upstream parameters are calculated according to the process mechanism of fly ash treatment. A time decay function is then introduced for correction, and the risk transmission coefficient is calculated, defined as follows:

[0027] ;

[0028] Among them, T i→j Δt represents the risk transmission coefficient from stage Ni to stage Nj, used to quantify the impact intensity of upstream anomalies on downstream anomalies; Δt represents the transmission time of fly ash from stage Ni to Nj; γ represents the system time constant; and γ represents the decay rate of the risk impact. The partial derivative of parameter x(j,l) of stage Nj with respect to parameter x(i,k) of stage Ni under the process mechanism reflects the causal strength based on the process mechanism, and e represents the natural constant.

[0029] The relationship between the defect rate of the final product in the fly ash treatment production line and the risk transmission coefficient from stage Ni to stage Nj was analyzed. A dynamic adjustment model was established. When the defect rate exceeds the set defect rate threshold, the risk transmission coefficient was optimized based on real-time monitoring data and process mechanism. Specific measures include: correcting the process parameters of stage Ni and injecting compensation actions in stage Nj.

[0030] A real-time monitoring system for fly ash processing production lines based on multi-sensor fusion includes a production line process analysis module, a data acquisition module, a risk assessment and calculation module, and a risk transmission module. The production line process analysis module is used to divide the fly ash processing production line into stages based on the fly ash processing process. The data acquisition module is used to collect data information from each stage of the fly ash processing production line through sensors. The risk assessment and calculation module is used to define a risk assessment function for each stage. The risk transmission module is used to establish a dynamic adjustment model by analyzing the relationship between the calculated risk transmission coefficient and the product defect rate.

[0031] The production line process analysis module includes a stage division unit and a parameter identification unit. The stage division unit is used to divide the entire fly ash processing production line into stages according to the fly ash processing process. The parameter identification unit is used to determine the input parameters that each stage receives from the upstream stage and the output parameters that are passed to the downstream stage.

[0032] The data acquisition module includes a data acquisition unit and a preprocessing unit. The data acquisition unit is used to install sensors at various locations on the fly ash processing production line to collect data information from each stage of the fly ash processing production line in real time. The preprocessing unit is used to clean the collected data and uses the Min-Max normalization algorithm to process the data information from each stage of the fly ash processing production line, and then combines the cleaned data from each stage into separate datasets.

[0033] The risk assessment calculation module includes a risk calculation unit and a correlation analysis unit. The risk calculation unit is used to calculate the safety risk assessment value of the current operating state of the current stage by comprehensively considering the weight, real-time value, target value and upper limit of the physical parameters monitored in the stage. The correlation analysis unit is used to extract the combination of upstream and downstream parameters based on the real-time parameters of each stage of the fly ash processing production line, and filter the parameter combinations whose gray correlation coefficient exceeds the set correlation strength threshold by gray correlation degree calculation.

[0034] The risk transmission module includes a partial derivative calculation unit, a coefficient correction unit, and a model building unit. The partial derivative calculation unit is used to calculate the partial derivatives of downstream parameters with respect to upstream parameters based on the parameter combinations whose gray correlation coefficients exceed the set correlation strength threshold, combined with the process mechanism of fly ash treatment. The coefficient correction unit is used to introduce a time decay function correction to calculate the risk transmission coefficient. The model building unit is used to analyze its relationship with the product defect rate and establish a dynamic adjustment model.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] 1. By fusion of multiple sensors to collect real-time data of multiple stages and parameters of the fly ash treatment production line, and combining grey relational analysis and process mechanism calculation, the risk transmission relationship between each stage is dynamically quantified. Unlike existing technologies that only focus on a single stage or parameter, this invention captures the correlation between upstream and downstream parameters of the production line, and realizes comprehensive risk assessment of the entire process from raw material input to product output.

[0037] 2. This invention constructs a dynamic adjustment model based on product defect rate feedback. Combining real-time monitoring data and process mechanisms, it injects compensation actions. Unlike the manual experience-based adjustment method in the prior art, this invention optimizes the risk transmission coefficient according to the real-time operating status of the production line and dynamically adjusts the production line operation to ensure the stability of the fly ash treatment process. Attached Figure Description

[0038] Figure 1 This is a flowchart illustrating the real-time monitoring method for fly ash treatment production lines based on multi-sensor fusion according to the present invention.

[0039] Figure 2 This is a schematic diagram of the structure of the real-time monitoring system for fly ash processing production line based on multi-sensor fusion according to the present invention. Detailed Implementation

[0040] 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.

[0041] In the embodiment: such as Figures 1-2 As shown, this invention provides a technical solution: a real-time monitoring method for fly ash processing lines based on multi-sensor fusion. The method includes the following steps:

[0042] Based on the fly ash treatment process, the fly ash treatment production line is divided into stages. After the division is completed, the input parameters of each stage that receive the output of the upstream stage and the output parameters that are transmitted to the downstream stage are obtained.

[0043] Data from each stage of the fly ash processing production line is collected by sensors, and the data from each stage is compiled into a dataset.

[0044] By comprehensively considering the weights, real-time values, target values, and upper limits of operational deviations of the physical parameters monitored during the phase, the safety risk assessment value of this phase under the current operational state is calculated.

[0045] Based on the real-time parameters of each stage of the fly ash treatment production line, upstream and downstream parameter combinations are extracted. Highly correlated parameter combinations are selected through gray relational degree calculation. The risk transmission coefficient is obtained by calculating the partial derivatives in combination with the process mechanism and introducing time decay correction. The relationship between the coefficient and the product defect rate is analyzed to establish a dynamic adjustment model.

[0046] Based on the fly ash treatment process, the fly ash treatment production line is divided into stages. After the division, the input parameters of each stage that receive the output from the upstream stage and the output parameters that are passed to the downstream stage are obtained. The specific steps include:

[0047] The fly ash treatment production line is divided into n stages according to the treatment process, denoted as N={N1,N2,...,Nn}, where N1,N2,...,Nn represent the corresponding 1st, 2nd,...,nth fly ash treatment production line stages, and n represents the total number of stages into which the fly ash treatment production line is divided.

[0048] After dividing the fly ash treatment production line into stages, each stage is marked with i. The input and output parameters of each stage are obtained. The input parameters of each stage are inherited from the output of the upstream stage, and the output parameters are passed to the downstream stage. Here, stage i represents the i-th fly ash treatment stage.

[0049] Specifically, the fly ash treatment production line is divided into five stages: feeding stage, water washing stage, stabilization treatment stage, high-temperature melting stage, and solidification and molding stage. Among them, the feeding stage N1 has the following input parameters: initial mass of fly ash (set as x(1,1)), initial moisture content (set as x(1,2)), and initial heavy metal content (set as x(1,3)); and output parameters: mass of fly ash entering the water washing stage (set as x(2,1)), moisture content (set as x(2,2)), and heavy metal content (set as x(2,3)).

[0050] Water washing stage N2: The input parameters are the output parameters x(2,1), x(2,2), x(2,3) of the feeding stage, as well as the water consumption for water washing (set as x(2,4)) and the water washing time (set as x(2,5)); the output parameters are the mass of fly ash after water washing (set as x(3,1)), moisture content (set as x(3,2)), and heavy metal content (set as x(3,3)). The various parameters of fly ash after water washing are used as the input parameters for the stabilization treatment stage.

[0051] Stabilization treatment stage N3: The input parameters include the output parameters of the water washing stage and the amount of stabilizer added (x(3,4)); the output parameter is the heavy metal stabilization rate of the fly ash after stabilization (set as x(4,1)), which is input to the high temperature melting stage;

[0052] High-temperature melting stage N4: The input parameters are the heavy metal stabilization rate of the stabilized fly ash (x(4,1), melting temperature (x(4,2)), melting time (x(4,3)), and heavy metal content of the material before melting (x(4,4)); the output parameters are the heavy metal content of the material after melting (x(5,1)) and material flowability (x(5,2)). These parameters are used in the solidification and molding stage.

[0053] The solidification and molding stage N5 has the following input parameters: the relevant parameters of the molten material x(5,1) and x(5,2), as well as the solidification temperature (set as x(5,3)) and pressure (set as x(5,4)); the output parameters are the compressive strength and heavy metal leaching concentration of the final product (solidified fly ash product).

[0054] Data from each stage of the fly ash treatment production line is collected using sensors, and the data from each stage is compiled into a dataset. The specific steps include:

[0055] Data information from each stage of the fly ash treatment production line is acquired through sensors. The acquired data information from each stage of the fly ash treatment production line is cleaned and preprocessed. When a data value is missing at a certain moment, a time series-based interpolation method is used to fill it in based on the data values ​​of the preceding and following moments. The data format is standardized and all data is stored in a unified database table structure.

[0056] The Min-Max normalization algorithm is used to process the data information of each stage of the fly ash processing production line and map the data to the [0,1] interval.

[0057] The cleaned data from each stage are grouped into datasets, denoted as {x1,x2,...,xn}. The dataset x1 from the first stage contains m1 parameters and is denoted as {x(1,1),x(1,2),...,x(1,m1)}. The dataset x2 from the second stage contains m2 parameters and is denoted as {x(2,1),x(2,2),...,x(2,m2)}, and so on. The dataset xn from the nth stage contains mn parameters and is denoted as {x(n,1),x(n,2),...,x(n,m)}. n Let x(1,1), x(1,2), ..., x(1,m1) represent the 1st, 2nd, ..., m1st parameters of the 1st stage, x(2,1), x(2,2), ..., x(2,m2) represent the 1st, 2nd, ..., m2nd parameters of the 2nd stage, and so on, x(n,1), x(n,2), ..., x(n,m) n ) represent the nth stage, 1, 2, ..., m respectively. nThere are several parameters, m1, m2, ..., m n These represent the total number of parameters in the 1st, 2nd, ..., nth stages, respectively.

[0058] Specifically, corresponding sensors are installed at each stage to collect data. For example, in the feeding stage, an electronic scale is used to measure the fly ash mass, a humidity sensor to measure the moisture content, and an atomic absorption spectrometer to detect the heavy metal content; in the washing stage, a flow sensor is used to monitor the water consumption for washing, and a timer records the washing time; in the stabilization treatment stage, a flow sensor is used to monitor the amount of stabilizer added; in the high-temperature melting stage, a temperature sensor is used to measure the melting temperature, and a timer records the melting time; in the curing and molding stage, a temperature sensor and a pressure sensor are used to obtain the curing temperature and pressure, a material testing machine is used to test the compressive strength of the product, and a chemical analyzer is used to detect the leaching concentration of heavy metals; after collecting data every minute for 1 hour, cleaning and preprocessing are performed. Based on the missing water consumption data at the 15th minute of the washing stage, and based on the water consumption of 500L at the 14th minute and 520L at the 16th minute, the missing value 510L is obtained through time series interpolation. Then, the Min-Max normalization algorithm is used to map the data to the [0,1] interval. After processing, the data from each stage are combined into a dataset.

[0059] By comprehensively considering the weights, real-time values, target values, and upper limits of operational deviations of the physical parameters monitored during the phase, the safety risk assessment value of this phase under the current operational state is calculated. The specific steps include:

[0060] A risk assessment function is defined for each stage, as shown below:

[0061] ;

[0062] Among them, R i This represents the safety risk assessment value for the i-th stage, reflecting the comprehensive degree of anomaly of all parameters in that stage. i w represents the total number of parameters monitored in the i-th stage. i,k Let x(i,k) represent the weight of the k-th parameter in the i-th stage. This weight is determined through training with historical data. Let x(i,k) represent the real-time measured value of the k-th parameter in the i-th stage. Let T(i,k) represent the target value of the k-th parameter in the i-th stage, which is the ideal state required by the process design. Let H(i,k) represent the upper limit of the operating deviation of the k-th parameter in the i-th stage, which is the maximum tolerable range of the actual value deviating from the target value.

[0063] Specifically, a risk assessment function is defined for each stage. For the feeding stage N1, m1=3 is set, and the weights are determined to be 0.3, 0.3, and 0.4 respectively through training with historical data. The target value T(1,1) is normalized to 0.5 and the upper limit of the operating deviation H(1,1) is normalized to 0.1. The target value T(1,2) is normalized to 0.4 and the upper limit of the operating deviation H(1,2) is normalized to 0.1. The target value T(1,3) is normalized to 0.5 and the upper limit of the operating deviation H(1,3) is normalized to 0.1. The normalized values ​​of x(1,1) are 0.55, x(1,2) are normalized to 0.45, and x(1,3) are normalized to 0.55. R1=0.5 is calculated. Similarly, R2=0.32, R3=0.35, R4=0.26 and R5=0.38 are calculated.

[0064] Based on real-time parameters at each stage of the fly ash treatment production line, upstream and downstream parameter combinations are extracted. Highly correlated parameter combinations are selected through grey relational analysis. Partial derivatives are calculated using the process mechanism, and a risk transmission coefficient is obtained by introducing time decay correction. The relationship between this coefficient and the product defect rate is analyzed to establish a dynamic adjustment model. Specific steps include:

[0065] Based on real-time parameter data of each stage of the fly ash treatment production line, parameter combinations between upstream and downstream stages are extracted. According to the grey relational degree, the grey relational coefficient γ(x(i,k),x(j,l)) of the parameter combinations between upstream and downstream stages is calculated. By setting a correlation strength threshold, parameter combinations whose grey relational coefficients exceed the set correlation strength threshold are filtered out. Here, γ(x(i,k),x(j,l)) represents the grey relational coefficient between parameter x(i,k) of stage Ni and parameter x(j,l) of stage Nj, which measures the statistical correlation strength between parameter x(i,k) of stage Ni and parameter x(j,l) of stage Nj.

[0066] The correlation coefficient is calculated using grey relational analysis. The formula for grey relational analysis is as follows:

[0067] ;

[0068] Where γ(x(i,k),x(j,l)) represents the statistical correlation strength between the real-time values ​​of the k-th parameter of stage Ni and the real-time values ​​of the l-th parameter of stage Nj, and ρ represents the resolution coefficient.

[0069] Based on the parameter combinations whose gray correlation coefficients exceed the set correlation strength threshold, the partial derivatives of downstream parameters with respect to upstream parameters are calculated according to the process mechanism of fly ash treatment. A time decay function is then introduced for correction, and the risk transmission coefficient is calculated, defined as follows:

[0070] ;

[0071] Among them, T i→j Δt represents the risk transmission coefficient from stage Ni to stage Nj, used to quantify the impact intensity of upstream anomalies on downstream anomalies; Δt represents the transmission time of fly ash from stage Ni to Nj; γ represents the system time constant; and γ represents the decay rate of the risk impact. The partial derivative of parameter x(j,l) of stage Nj with respect to parameter x(i,k) of stage Ni under the process mechanism reflects the causal strength based on the process mechanism, and e represents the natural constant.

[0072] The relationship between the defect rate of the final product in the fly ash treatment production line and the risk transmission coefficient from stage Ni to stage Nj was analyzed. A dynamic adjustment model was established. When the defect rate exceeds the set defect rate threshold, the risk transmission coefficient was optimized based on real-time monitoring data and process mechanism. Specific measures include: correcting the process parameters of stage Ni and injecting compensation actions in stage Nj.

[0073] Specifically, parameter combinations between upstream and downstream stages are extracted. For example, the water consumption in the washing stage and the heavy metal stabilization rate in the stabilization treatment stage are calculated, and their grey correlation coefficient is 0.75. A correlation strength threshold of 0.6 is set, and this parameter combination is filtered out because its correlation coefficient exceeds the threshold. Partial derivatives are calculated and time decay correction is introduced: Based on the process mechanism, it is known that for every unit increase in water consumption during washing, the heavy metal stabilization rate in the stabilization treatment stage increases by 0.4 units. The transfer time of fly ash from the washing stage to the stabilization treatment stage is 0.3. Therefore, the risk transfer coefficient T... 2→3 =0.22224; Continuously monitor the defect rate of the final product, set the defect rate threshold to 5%, and when the defect rate reaches 6%, optimize the risk transmission coefficient based on real-time monitoring data and process mechanism.

[0074] The real-time monitoring system for fly ash processing production lines based on multi-sensor fusion includes a production line process analysis module, a data acquisition module, a risk assessment and calculation module, and a risk transmission module. The production line process analysis module is used to divide the fly ash processing production line into stages based on the fly ash processing process; the data acquisition module is used to collect data information from each stage of the fly ash processing production line through sensors; the risk assessment and calculation module is used to define a risk assessment function for each stage; and the risk transmission module is used to establish a dynamic adjustment model by analyzing the relationship between the calculated risk transmission coefficient and the product defect rate.

[0075] The production line process analysis module includes a stage division unit and a parameter identification unit. The stage division unit is used to divide the entire fly ash processing production line into stages according to the fly ash processing process. The parameter identification unit is used to determine the input parameters that each stage receives from the upstream stage and the output parameters that are passed to the downstream stage.

[0076] The data acquisition module includes a data acquisition unit and a preprocessing unit. The data acquisition unit is used to install sensors at various locations on the fly ash processing production line to collect data information at each stage of the fly ash processing production line in real time. The preprocessing unit is used to clean the collected data and uses the Min-Max normalization algorithm to process the data information at each stage of the fly ash processing production line, and then combines the cleaned data from each stage into separate datasets.

[0077] The risk assessment calculation module includes a risk calculation unit and a correlation analysis unit. The risk calculation unit is used to calculate the safety risk assessment value of the current operating state of the stage by comprehensively considering the weight, real-time value, target value and upper limit of the physical parameters monitored in the stage. The correlation analysis unit is used to extract the combination of upstream and downstream parameters based on the real-time parameters of each stage of the fly ash processing production line, and to filter the parameter combination whose gray correlation coefficient exceeds the set correlation strength threshold by gray correlation degree calculation.

[0078] The risk transmission module includes a partial derivative calculation unit, a coefficient correction unit, and a model building unit. The partial derivative calculation unit is used to calculate the partial derivatives of downstream parameters with respect to upstream parameters based on the parameter combinations whose gray correlation coefficients exceed the set correlation strength threshold, combined with the process mechanism of fly ash treatment. The coefficient correction unit is used to introduce a time decay function correction to calculate the risk transmission coefficient. The model building unit is used to analyze its relationship with the product defect rate and establish a dynamic adjustment model.

[0079] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A real-time monitoring method for fly ash treatment production lines based on multi-sensor fusion, characterized in that: The method includes the following steps: Based on the fly ash treatment process, the fly ash treatment production line is divided into stages. After the division is completed, the input parameters of each stage that receive the output of the upstream stage and the output parameters that are transmitted to the downstream stage are obtained. Data from each stage of the fly ash processing production line is collected by sensors, and the data from each stage is compiled into a dataset. By comprehensively considering the weights, real-time values, target values, and upper limits of operational deviations of the physical parameters monitored during the phase, the safety risk assessment value of this phase under the current operational state is calculated. A risk assessment function is defined for each stage, as shown below: ； Among them, R i This represents the safety risk assessment value for the i-th stage, reflecting the comprehensive degree of anomaly of all parameters in that stage. i w represents the total number of parameters monitored in the i-th stage. i,k Let x(i,k) represent the weight of the k-th parameter in the i-th stage, which is determined through training with historical data. Let x(i,k) represent the real-time measurement value of the k-th parameter in the i-th stage. Let T(i,k) represent the target value of the k-th parameter in the i-th stage, which is the ideal state required by the process design. Let H(i,k) represent the upper limit of the operating deviation of the k-th parameter in the i-th stage, which is the maximum tolerable range of the actual value deviating from the target value. Based on real-time parameters at each stage of the fly ash treatment production line, upstream and downstream parameter combinations are extracted. Highly correlated parameter combinations are selected through gray relational degree calculation. Partial derivatives are calculated in combination with process mechanism and time decay correction is introduced to obtain risk transmission coefficient. The relationship between the coefficient and product defect rate is analyzed to establish a dynamic adjustment model. Based on real-time parameter data of each stage of the fly ash treatment production line, parameter combinations between upstream and downstream stages are extracted. According to the grey relational degree, the grey relational coefficient γ(x(i,k),x(j,l)) of the parameter combinations between upstream and downstream stages is calculated. By setting a correlation strength threshold, parameter combinations whose grey relational coefficients exceed the set correlation strength threshold are filtered out. Here, γ(x(i,k),x(j,l)) represents the grey relational coefficient between parameter x(i,k) of stage Ni and parameter x(j,l) of stage Nj, which measures the statistical correlation strength between parameter x(i,k) of stage Ni and parameter x(j,l) of stage Nj. Based on the parameter combinations whose gray correlation coefficients exceed the set correlation strength threshold, the partial derivatives of downstream parameters with respect to upstream parameters are calculated according to the process mechanism of fly ash treatment. A time decay function is then introduced for correction, and the risk transmission coefficient is calculated, defined as follows: ； Among them, T i→j denoted as the risk transmission coefficient from stage Ni to stage Nj, used to quantify the impact intensity of upstream anomalies on downstream anomalies; Δt represents the transmission time of fly ash from stage Ni to Nj; γ represents the system time constant; γ represents the decay rate of the risk impact; ∂x(i,k) / ∂x(j,l) represents the partial derivative of the parameter x(j,l) of stage Nj with respect to the parameter x(i,k) of stage Ni based on the process mechanism, reflecting the causal strength based on the process mechanism; e represents the natural constant. The relationship between the defect rate of the final product in the fly ash treatment production line and the risk transmission coefficient from stage Ni to stage Nj was analyzed. A dynamic adjustment model was established. When the defect rate exceeds the set defect rate threshold, the risk transmission coefficient was optimized based on real-time monitoring data and process mechanism. Specific measures include: correcting the process parameters of stage Ni and injecting compensation actions in stage Nj.

2. The real-time monitoring method for fly ash treatment production line based on multi-sensor fusion according to claim 1, characterized in that: Based on the fly ash treatment process, the fly ash treatment production line is divided into stages. After the division, the input parameters of each stage that receive the output from the upstream stage and the output parameters that are passed to the downstream stage are obtained. The specific steps include: The fly ash treatment production line is divided into n stages according to the treatment process, denoted as N={N1,N2,...,Nn}, where N1,N2,...,Nn represent the corresponding 1st, 2nd,...,nth fly ash treatment production line stages, and n represents the total number of stages into which the fly ash treatment production line is divided. After dividing the fly ash treatment production line into stages, each stage is marked with i. The input and output parameters of each stage are obtained. The input parameters of each stage are inherited from the output of the upstream stage, and the output parameters are passed to the downstream stage. Here, stage i represents the i-th fly ash treatment stage.

3. The real-time monitoring method for fly ash treatment production line based on multi-sensor fusion according to claim 2, characterized in that: Data from each stage of the fly ash treatment production line is collected using sensors, and the data from each stage is compiled into a dataset. The specific steps include: Data information from each stage of the fly ash treatment production line is acquired through sensors. The acquired data information from each stage of the fly ash treatment production line is cleaned and preprocessed. When a data value is missing at a certain moment, a time series-based interpolation method is used to fill it in based on the data values ​​of the preceding and following moments. The data format is standardized and all data is stored in a unified database table structure. The Min-Max normalization algorithm is used to process the data information of each stage of the fly ash processing production line and map the data to the [0,1] interval. The cleaned data from each stage are grouped into datasets, denoted as {x1,x2,...,xn}. The dataset x1 from the first stage contains m1 parameters and is denoted as {x(1,1),x(1,2),...,x(1,m1)}. The dataset x2 from the second stage contains m2 parameters and is denoted as {x(2,1),x(2,2),...,x(2,m2)}, and so on. The dataset xn from the nth stage contains mn parameters and is denoted as {x(n,1),x(n,2),...,x(n,m)}. n Let x(1,1), x(1,2), ..., x(1,m1) represent the 1st, 2nd, ..., m1st parameters of the 1st stage, x(2,1), x(2,2), ..., x(2,m2) represent the 1st, 2nd, ..., m2nd parameters of the 2nd stage, and so on, x(n,1), x(n,2), ..., x(n,m) n ) represent the nth stage, 1, 2, ..., m respectively. n There are several parameters, m1, m2, ..., m n These represent the total number of parameters in the 1st, 2nd, ..., nth stages, respectively.

4. A real-time monitoring system for fly ash processing lines based on multi-sensor fusion, applied to the real-time monitoring method for fly ash processing lines based on multi-sensor fusion as described in any one of claims 1-3, characterized in that: The system includes a production line process analysis module, a data acquisition module, a risk assessment calculation module, and a risk transmission module. The production line process analysis module is used to divide the fly ash processing production line into stages based on the fly ash processing process. The data acquisition module is used to collect data information of each stage of the fly ash processing production line through sensors. The risk assessment calculation module is used to define a risk assessment function for each stage. The risk transmission module is used to establish a dynamic adjustment model by analyzing the relationship between the calculated risk transmission coefficient and the product defect rate.

5. The real-time monitoring system for fly ash processing lines based on multi-sensor fusion according to claim 4, characterized in that: The production line process analysis module includes a stage division unit and a parameter identification unit. The stage division unit is used to divide the entire fly ash processing production line into stages according to the fly ash processing process. The parameter identification unit is used to determine the input parameters that each stage receives from the upstream stage and the output parameters that are passed to the downstream stage.

6. The real-time monitoring system for fly ash treatment production line based on multi-sensor fusion according to claim 5, characterized in that: The data acquisition module includes a data acquisition unit and a preprocessing unit. The data acquisition unit is used to install sensors at various locations on the fly ash processing production line to collect data information from each stage of the fly ash processing production line in real time. The preprocessing unit is used to clean the collected data and uses the Min-Max normalization algorithm to process the data information from each stage of the fly ash processing production line, and then combines the cleaned data from each stage into separate datasets.

7. The real-time monitoring system for fly ash processing line based on multi-sensor fusion according to claim 6, characterized in that: The risk assessment calculation module includes a risk calculation unit and a correlation analysis unit. The risk calculation unit is used to calculate the safety risk assessment value of the current operating state of the current stage by comprehensively considering the weight, real-time value, target value and upper limit of the physical parameters monitored in the stage. The correlation analysis unit is used to extract upstream and downstream parameter combinations based on real-time parameters at each stage of the fly ash processing production line, and to filter parameter combinations whose gray correlation coefficient exceeds the set correlation strength threshold through gray correlation degree calculation.

8. The real-time monitoring system for fly ash treatment production line based on multi-sensor fusion according to claim 7, characterized in that: The risk transmission module includes a partial derivative calculation unit, a coefficient correction unit, and a model building unit. The partial derivative calculation unit is used to calculate the partial derivatives of downstream parameters with respect to upstream parameters based on the parameter combinations whose gray correlation coefficients exceed the set correlation strength threshold, combined with the process mechanism of fly ash treatment. The coefficient correction unit is used to introduce a time decay function correction to calculate the risk transmission coefficient. The model building unit is used to analyze its relationship with the product defect rate and establish a dynamic adjustment model.

Images